From f4e5721b53e8e0374d5f7298120193dc2e298dc4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rok=20Gomi=C5=A1=C4=8Dek?= <rok.gomiscek@gmail.com>
Date: Tue, 2 Dec 2025 14:41:39 +0100
Subject: [PATCH] Klasifikacija - posodobitev

---
 notebooks/data/sportniki.tab                  |    4 +-
 notebooks/{podatki => data}/titanic-test.tab  | 2208 ++++++++--------
 .../{podatki => data}/titanic-training.tab    | 2206 ++++++++--------
 notebooks/slike/type12_error.jpeg             |  Bin 9628 -> 0 bytes
 .../src/107-1_nadzorovano_naivniBayes.ipynb   | 2313 ++++++++++++-----
 notebooks/src/207-1.ipynb                     |  767 +++++-
 6 files changed, 4563 insertions(+), 2935 deletions(-)
 rename notebooks/{podatki => data}/titanic-test.tab (95%)
 rename notebooks/{podatki => data}/titanic-training.tab (95%)
 delete mode 100644 notebooks/slike/type12_error.jpeg

diff --git a/notebooks/data/sportniki.tab b/notebooks/data/sportniki.tab
index 9d60a71..f626dc5 100644
--- a/notebooks/data/sportniki.tab
+++ b/notebooks/data/sportniki.tab
@@ -1,6 +1,4 @@
 visina	sport
-d	d
-	class
 visok	kosarka
 visok	kosarka
 visok	kosarka
@@ -20,4 +18,4 @@ nizek	gimnastika
 nizek	gimnastika
 nizek	gimnastika
 srednji	gimnastika
-srednji	gimnastika
+srednji	gimnastika
\ No newline at end of file
diff --git a/notebooks/podatki/titanic-test.tab b/notebooks/data/titanic-test.tab
similarity index 95%
rename from notebooks/podatki/titanic-test.tab
rename to notebooks/data/titanic-test.tab
index 59f33b1..7e8962d 100644
--- a/notebooks/podatki/titanic-test.tab
+++ b/notebooks/data/titanic-test.tab
@@ -1,1104 +1,1104 @@
-status	age	sex	survived
-discrete	discrete	discrete	discrete
-			class
-crew	adult	male	yes
-first	adult	male	no
-third	adult	male	no
-first	child	male	yes
-third	child	female	yes
-first	adult	male	yes
-crew	adult	male	no
-first	adult	female	yes
-third	child	male	yes
-crew	adult	male	yes
-third	child	male	no
-third	adult	male	no
-second	adult	female	yes
-second	child	female	yes
-first	adult	female	yes
-first	adult	female	yes
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-second	adult	female	yes
-first	adult	male	no
-second	adult	male	no
-third	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-second	adult	male	no
-second	adult	female	no
-crew	adult	male	no
-third	child	male	yes
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-third	adult	male	no
-third	adult	male	yes
-first	adult	female	yes
-second	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-second	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-second	adult	male	yes
-crew	adult	male	no
-first	adult	male	yes
-third	adult	male	no
-first	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-third	adult	male	no
-first	adult	male	no
-first	child	male	yes
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	yes
-second	adult	female	yes
-first	adult	male	no
-first	adult	male	no
-second	adult	female	yes
-first	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	yes
-crew	adult	male	yes
-third	child	male	no
-third	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-second	adult	female	no
-third	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	male	yes
-first	adult	male	yes
-crew	adult	male	yes
-second	adult	male	no
-first	adult	male	yes
-crew	adult	male	no
-second	adult	female	yes
-first	adult	female	yes
-third	adult	male	no
-second	adult	female	yes
-first	adult	female	yes
-third	adult	male	no
-crew	adult	male	yes
-third	adult	female	no
-crew	adult	male	yes
-third	child	female	yes
-third	adult	male	no
-crew	adult	female	yes
-crew	adult	male	yes
-first	adult	female	yes
-third	adult	male	yes
-crew	adult	male	yes
-first	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-second	child	male	yes
-crew	adult	female	yes
-third	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	yes
-first	adult	female	yes
-crew	adult	male	yes
-third	adult	female	no
-crew	adult	male	no
-third	child	female	no
-first	adult	male	no
-first	adult	male	yes
-third	adult	female	no
-first	adult	male	no
-second	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-third	child	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	child	male	no
-third	adult	male	yes
-second	adult	male	no
-third	child	male	yes
-first	adult	male	no
-first	adult	male	yes
-crew	adult	male	no
-second	adult	female	yes
-third	adult	male	no
-first	adult	male	yes
-third	adult	male	yes
-first	adult	female	yes
-third	adult	male	no
-crew	adult	male	yes
-third	adult	male	yes
-first	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-second	adult	male	yes
-first	adult	male	yes
-crew	adult	male	no
-first	adult	male	no
-second	adult	male	no
-third	adult	female	yes
-first	adult	female	yes
-first	adult	female	yes
-third	adult	female	yes
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-third	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-first	adult	female	yes
-third	adult	male	yes
-crew	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-second	child	female	yes
-crew	adult	male	yes
-third	adult	female	yes
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-third	adult	male	yes
-third	child	female	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-second	adult	male	no
-crew	adult	male	yes
-crew	adult	male	yes
-third	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-first	adult	male	no
-first	adult	male	yes
-first	adult	male	no
-first	adult	female	yes
-first	adult	male	yes
-second	adult	female	yes
-first	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-first	adult	female	yes
-first	adult	male	no
-third	adult	male	no
-third	adult	male	no
-first	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-first	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-second	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-crew	adult	male	yes
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-second	adult	male	no
-first	adult	male	no
-second	adult	female	no
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	female	no
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-second	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-first	adult	female	yes
-crew	adult	male	no
-crew	adult	female	yes
-first	adult	male	no
-crew	adult	male	yes
-first	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-first	adult	male	yes
-crew	adult	male	no
-third	adult	female	yes
-third	adult	female	yes
-crew	adult	male	yes
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	child	male	yes
-first	adult	male	no
-third	child	female	no
-third	adult	male	no
-crew	adult	male	no
-third	child	male	no
-first	adult	female	yes
-first	adult	male	no
-third	adult	female	yes
-second	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-third	child	male	no
-third	child	male	no
-crew	adult	male	no
-crew	adult	male	yes
-first	adult	male	yes
-second	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-third	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-third	adult	male	no
-second	adult	female	yes
-third	adult	male	no
-first	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-second	adult	female	yes
-third	adult	female	no
-crew	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	female	yes
-third	adult	male	no
-second	adult	male	no
-third	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-second	adult	male	no
-third	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-third	child	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-third	adult	male	no
-first	adult	male	no
-third	adult	female	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-first	adult	male	no
-first	adult	male	no
-third	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	female	no
-second	adult	female	yes
-crew	adult	male	no
-second	adult	male	no
-second	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-first	adult	male	no
-crew	adult	male	yes
-third	adult	female	no
-first	adult	male	yes
-third	adult	male	no
-first	adult	male	no
-first	adult	male	no
-third	adult	female	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-third	adult	female	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	yes
-second	adult	female	yes
-second	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	yes
-crew	adult	male	no
-third	adult	female	no
-first	adult	male	yes
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-third	child	male	no
-crew	adult	male	no
-third	adult	male	no
-second	adult	male	no
-third	adult	male	no
-third	adult	female	yes
-third	adult	male	no
-second	child	male	yes
-third	adult	male	no
-third	adult	male	yes
-first	adult	female	yes
-third	adult	male	no
-crew	adult	female	yes
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	female	no
-third	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-third	adult	female	no
-third	adult	female	no
-crew	adult	male	yes
-crew	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-second	adult	male	no
-first	adult	female	yes
-first	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-second	adult	male	no
-first	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	child	male	yes
-first	adult	male	no
-third	child	female	yes
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-first	adult	female	yes
-crew	adult	male	no
-crew	adult	male	yes
-first	adult	male	yes
-crew	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-third	adult	male	no
-second	adult	male	no
-crew	adult	male	yes
-third	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-second	adult	male	no
-third	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-first	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-second	child	female	yes
-crew	adult	male	no
-first	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-second	adult	female	yes
-third	child	male	no
-crew	adult	male	yes
-crew	adult	male	no
-first	adult	male	yes
-second	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-third	adult	female	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-first	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-second	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-second	adult	female	yes
-crew	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	child	male	yes
-third	adult	female	no
-crew	adult	male	no
-third	adult	female	yes
-second	adult	female	yes
-first	adult	male	yes
-second	adult	male	no
-third	child	male	no
-third	child	female	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-second	adult	male	no
-second	adult	male	no
-first	adult	female	yes
-third	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-first	adult	male	no
-third	adult	male	no
-third	adult	male	no
-third	adult	female	yes
-crew	adult	male	no
-third	child	female	no
-third	adult	male	no
-third	adult	female	yes
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-first	adult	male	no
-third	adult	female	yes
-crew	adult	male	no
-crew	adult	female	no
-third	adult	male	no
-second	adult	female	no
-crew	adult	male	yes
-third	adult	male	no
-third	adult	female	no
-third	child	female	yes
-crew	adult	male	no
-third	child	male	no
-second	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-third	adult	female	yes
-crew	adult	male	no
-third	adult	female	yes
-third	adult	male	no
-second	adult	female	yes
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-third	adult	male	no
-first	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-first	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-second	adult	male	no
-first	adult	female	yes
-third	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-third	adult	male	no
-first	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-third	adult	male	no
-third	adult	female	no
-third	adult	female	yes
-first	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	yes
-second	adult	male	no
-third	adult	male	no
-third	adult	female	yes
-third	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-first	adult	female	yes
-first	adult	male	no
-third	adult	female	no
-third	adult	female	no
-second	adult	female	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-third	adult	female	yes
-third	adult	female	no
-second	adult	male	no
-third	adult	male	no
-third	adult	male	yes
-second	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-first	child	male	yes
-first	adult	male	yes
-third	adult	male	yes
-first	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-second	adult	female	yes
-second	adult	male	no
-third	adult	male	no
-second	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	female	no
-first	adult	female	yes
-crew	adult	male	yes
-third	adult	female	yes
-third	adult	male	no
-crew	adult	male	yes
-second	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	child	female	no
-third	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-second	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	yes
-first	adult	male	no
-crew	adult	male	yes
-third	child	male	no
-crew	adult	female	yes
-third	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-first	adult	female	yes
-third	adult	female	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	child	female	yes
-first	adult	female	yes
-first	adult	female	yes
-second	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-crew	adult	male	yes
-third	adult	male	no
-second	adult	male	no
-first	adult	male	no
-third	child	male	yes
-third	adult	female	no
-first	adult	male	yes
-crew	adult	male	no
-third	adult	female	yes
-crew	adult	female	yes
-second	adult	female	yes
-crew	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-first	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-first	adult	female	yes
-first	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	female	yes
-third	adult	male	no
-third	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-third	adult	female	no
-third	adult	female	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	child	female	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-third	adult	female	yes
-third	child	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-second	child	male	yes
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-first	adult	male	yes
-first	adult	male	no
-second	adult	male	no
-third	adult	female	no
-first	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-third	child	male	no
-first	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	yes
-first	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-first	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-second	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	yes
-second	adult	male	no
-crew	adult	male	yes
-first	adult	female	yes
-third	adult	female	yes
-third	adult	female	no
-second	child	female	yes
-first	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-first	adult	female	yes
-third	adult	male	no
-second	adult	male	no
-first	adult	male	no
-third	adult	female	no
-first	adult	female	yes
-crew	adult	male	yes
-third	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-first	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-third	child	female	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-third	adult	male	yes
-third	child	male	no
-first	adult	female	no
-third	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	yes
-third	adult	male	no
-third	child	male	no
-crew	adult	male	no
-third	adult	male	no
-third	child	female	no
-crew	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-third	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-third	adult	male	no
-second	adult	female	yes
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-second	adult	male	no
-third	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	yes
-first	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-second	adult	male	no
-second	adult	male	no
-third	adult	male	no
-third	adult	male	no
-third	adult	female	yes
-third	adult	male	no
-second	adult	male	yes
-third	adult	male	no
-third	child	male	no
-second	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-third	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	yes
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-second	child	female	yes
-third	child	female	yes
-crew	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-third	child	male	no
-crew	adult	male	no
-second	adult	female	yes
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-second	adult	male	no
-third	adult	male	no
-first	adult	male	yes
-second	adult	male	no
-third	adult	female	no
-third	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-second	adult	male	yes
-third	child	male	yes
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-third	adult	female	yes
-third	child	female	yes
-third	adult	male	no
+status	age	sex	survived
+discrete	discrete	discrete	discrete
+			class
+crew	adult	male	yes
+first	adult	male	no
+third	adult	male	no
+first	child	male	yes
+third	child	female	yes
+first	adult	male	yes
+crew	adult	male	no
+first	adult	female	yes
+third	child	male	yes
+crew	adult	male	yes
+third	child	male	no
+third	adult	male	no
+second	adult	female	yes
+second	child	female	yes
+first	adult	female	yes
+first	adult	female	yes
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+second	adult	female	yes
+first	adult	male	no
+second	adult	male	no
+third	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+second	adult	male	no
+second	adult	female	no
+crew	adult	male	no
+third	child	male	yes
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+third	adult	male	no
+third	adult	male	yes
+first	adult	female	yes
+second	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+second	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+second	adult	male	yes
+crew	adult	male	no
+first	adult	male	yes
+third	adult	male	no
+first	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+third	adult	male	no
+first	adult	male	no
+first	child	male	yes
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	yes
+second	adult	female	yes
+first	adult	male	no
+first	adult	male	no
+second	adult	female	yes
+first	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	yes
+crew	adult	male	yes
+third	child	male	no
+third	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+second	adult	female	no
+third	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	male	yes
+first	adult	male	yes
+crew	adult	male	yes
+second	adult	male	no
+first	adult	male	yes
+crew	adult	male	no
+second	adult	female	yes
+first	adult	female	yes
+third	adult	male	no
+second	adult	female	yes
+first	adult	female	yes
+third	adult	male	no
+crew	adult	male	yes
+third	adult	female	no
+crew	adult	male	yes
+third	child	female	yes
+third	adult	male	no
+crew	adult	female	yes
+crew	adult	male	yes
+first	adult	female	yes
+third	adult	male	yes
+crew	adult	male	yes
+first	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+second	child	male	yes
+crew	adult	female	yes
+third	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	yes
+first	adult	female	yes
+crew	adult	male	yes
+third	adult	female	no
+crew	adult	male	no
+third	child	female	no
+first	adult	male	no
+first	adult	male	yes
+third	adult	female	no
+first	adult	male	no
+second	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+third	child	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	child	male	no
+third	adult	male	yes
+second	adult	male	no
+third	child	male	yes
+first	adult	male	no
+first	adult	male	yes
+crew	adult	male	no
+second	adult	female	yes
+third	adult	male	no
+first	adult	male	yes
+third	adult	male	yes
+first	adult	female	yes
+third	adult	male	no
+crew	adult	male	yes
+third	adult	male	yes
+first	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+second	adult	male	yes
+first	adult	male	yes
+crew	adult	male	no
+first	adult	male	no
+second	adult	male	no
+third	adult	female	yes
+first	adult	female	yes
+first	adult	female	yes
+third	adult	female	yes
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+third	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+first	adult	female	yes
+third	adult	male	yes
+crew	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+second	child	female	yes
+crew	adult	male	yes
+third	adult	female	yes
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+third	adult	male	yes
+third	child	female	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+second	adult	male	no
+crew	adult	male	yes
+crew	adult	male	yes
+third	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+first	adult	male	no
+first	adult	male	yes
+first	adult	male	no
+first	adult	female	yes
+first	adult	male	yes
+second	adult	female	yes
+first	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+first	adult	female	yes
+first	adult	male	no
+third	adult	male	no
+third	adult	male	no
+first	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+first	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+second	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+crew	adult	male	yes
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+second	adult	male	no
+first	adult	male	no
+second	adult	female	no
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	female	no
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+second	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+first	adult	female	yes
+crew	adult	male	no
+crew	adult	female	yes
+first	adult	male	no
+crew	adult	male	yes
+first	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+first	adult	male	yes
+crew	adult	male	no
+third	adult	female	yes
+third	adult	female	yes
+crew	adult	male	yes
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	child	male	yes
+first	adult	male	no
+third	child	female	no
+third	adult	male	no
+crew	adult	male	no
+third	child	male	no
+first	adult	female	yes
+first	adult	male	no
+third	adult	female	yes
+second	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+third	child	male	no
+third	child	male	no
+crew	adult	male	no
+crew	adult	male	yes
+first	adult	male	yes
+second	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+third	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+third	adult	male	no
+second	adult	female	yes
+third	adult	male	no
+first	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+second	adult	female	yes
+third	adult	female	no
+crew	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	female	yes
+third	adult	male	no
+second	adult	male	no
+third	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+second	adult	male	no
+third	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+third	child	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+third	adult	male	no
+first	adult	male	no
+third	adult	female	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+first	adult	male	no
+first	adult	male	no
+third	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	female	no
+second	adult	female	yes
+crew	adult	male	no
+second	adult	male	no
+second	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+first	adult	male	no
+crew	adult	male	yes
+third	adult	female	no
+first	adult	male	yes
+third	adult	male	no
+first	adult	male	no
+first	adult	male	no
+third	adult	female	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+third	adult	female	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	yes
+second	adult	female	yes
+second	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	yes
+crew	adult	male	no
+third	adult	female	no
+first	adult	male	yes
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+third	child	male	no
+crew	adult	male	no
+third	adult	male	no
+second	adult	male	no
+third	adult	male	no
+third	adult	female	yes
+third	adult	male	no
+second	child	male	yes
+third	adult	male	no
+third	adult	male	yes
+first	adult	female	yes
+third	adult	male	no
+crew	adult	female	yes
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	female	no
+third	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+third	adult	female	no
+third	adult	female	no
+crew	adult	male	yes
+crew	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+second	adult	male	no
+first	adult	female	yes
+first	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+second	adult	male	no
+first	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	child	male	yes
+first	adult	male	no
+third	child	female	yes
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+first	adult	female	yes
+crew	adult	male	no
+crew	adult	male	yes
+first	adult	male	yes
+crew	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+third	adult	male	no
+second	adult	male	no
+crew	adult	male	yes
+third	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+second	adult	male	no
+third	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+first	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+second	child	female	yes
+crew	adult	male	no
+first	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+second	adult	female	yes
+third	child	male	no
+crew	adult	male	yes
+crew	adult	male	no
+first	adult	male	yes
+second	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+third	adult	female	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+first	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+second	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+second	adult	female	yes
+crew	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	child	male	yes
+third	adult	female	no
+crew	adult	male	no
+third	adult	female	yes
+second	adult	female	yes
+first	adult	male	yes
+second	adult	male	no
+third	child	male	no
+third	child	female	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+second	adult	male	no
+second	adult	male	no
+first	adult	female	yes
+third	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+first	adult	male	no
+third	adult	male	no
+third	adult	male	no
+third	adult	female	yes
+crew	adult	male	no
+third	child	female	no
+third	adult	male	no
+third	adult	female	yes
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+first	adult	male	no
+third	adult	female	yes
+crew	adult	male	no
+crew	adult	female	no
+third	adult	male	no
+second	adult	female	no
+crew	adult	male	yes
+third	adult	male	no
+third	adult	female	no
+third	child	female	yes
+crew	adult	male	no
+third	child	male	no
+second	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+third	adult	female	yes
+crew	adult	male	no
+third	adult	female	yes
+third	adult	male	no
+second	adult	female	yes
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+third	adult	male	no
+first	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+first	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+second	adult	male	no
+first	adult	female	yes
+third	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+third	adult	male	no
+first	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+third	adult	male	no
+third	adult	female	no
+third	adult	female	yes
+first	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	yes
+second	adult	male	no
+third	adult	male	no
+third	adult	female	yes
+third	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+first	adult	female	yes
+first	adult	male	no
+third	adult	female	no
+third	adult	female	no
+second	adult	female	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+third	adult	female	yes
+third	adult	female	no
+second	adult	male	no
+third	adult	male	no
+third	adult	male	yes
+second	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+first	child	male	yes
+first	adult	male	yes
+third	adult	male	yes
+first	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+second	adult	female	yes
+second	adult	male	no
+third	adult	male	no
+second	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	female	no
+first	adult	female	yes
+crew	adult	male	yes
+third	adult	female	yes
+third	adult	male	no
+crew	adult	male	yes
+second	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	child	female	no
+third	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+second	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	yes
+first	adult	male	no
+crew	adult	male	yes
+third	child	male	no
+crew	adult	female	yes
+third	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+first	adult	female	yes
+third	adult	female	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	child	female	yes
+first	adult	female	yes
+first	adult	female	yes
+second	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+crew	adult	male	yes
+third	adult	male	no
+second	adult	male	no
+first	adult	male	no
+third	child	male	yes
+third	adult	female	no
+first	adult	male	yes
+crew	adult	male	no
+third	adult	female	yes
+crew	adult	female	yes
+second	adult	female	yes
+crew	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+first	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+first	adult	female	yes
+first	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	female	yes
+third	adult	male	no
+third	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+third	adult	female	no
+third	adult	female	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	child	female	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+third	adult	female	yes
+third	child	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+second	child	male	yes
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+first	adult	male	yes
+first	adult	male	no
+second	adult	male	no
+third	adult	female	no
+first	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+third	child	male	no
+first	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	yes
+first	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+first	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+second	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	yes
+second	adult	male	no
+crew	adult	male	yes
+first	adult	female	yes
+third	adult	female	yes
+third	adult	female	no
+second	child	female	yes
+first	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+first	adult	female	yes
+third	adult	male	no
+second	adult	male	no
+first	adult	male	no
+third	adult	female	no
+first	adult	female	yes
+crew	adult	male	yes
+third	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+first	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+third	child	female	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+third	adult	male	yes
+third	child	male	no
+first	adult	female	no
+third	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	yes
+third	adult	male	no
+third	child	male	no
+crew	adult	male	no
+third	adult	male	no
+third	child	female	no
+crew	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+third	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+third	adult	male	no
+second	adult	female	yes
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+second	adult	male	no
+third	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	yes
+first	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+second	adult	male	no
+second	adult	male	no
+third	adult	male	no
+third	adult	male	no
+third	adult	female	yes
+third	adult	male	no
+second	adult	male	yes
+third	adult	male	no
+third	child	male	no
+second	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+third	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	yes
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+second	child	female	yes
+third	child	female	yes
+crew	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+third	child	male	no
+crew	adult	male	no
+second	adult	female	yes
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+second	adult	male	no
+third	adult	male	no
+first	adult	male	yes
+second	adult	male	no
+third	adult	female	no
+third	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+second	adult	male	yes
+third	child	male	yes
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+third	adult	female	yes
+third	child	female	yes
+third	adult	male	no
diff --git a/notebooks/podatki/titanic-training.tab b/notebooks/data/titanic-training.tab
similarity index 95%
rename from notebooks/podatki/titanic-training.tab
rename to notebooks/data/titanic-training.tab
index 26b843d..e75123e 100644
--- a/notebooks/podatki/titanic-training.tab
+++ b/notebooks/data/titanic-training.tab
@@ -1,1103 +1,1103 @@
-status	age	sex	survived
-discrete	discrete	discrete	discrete
-			class
-third	adult	male	no
-second	adult	female	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-third	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	female	yes
-crew	adult	male	yes
-first	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	yes
-first	adult	female	yes
-second	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-first	adult	male	yes
-third	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-crew	adult	male	yes
-first	adult	female	yes
-third	adult	female	yes
-crew	adult	male	no
-third	adult	male	yes
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-third	adult	female	no
-crew	adult	male	yes
-first	adult	male	no
-first	adult	female	yes
-first	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	female	no
-second	adult	male	no
-second	adult	female	yes
-third	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-first	adult	female	yes
-third	adult	female	no
-crew	adult	male	yes
-crew	adult	male	no
-second	adult	female	yes
-first	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	female	yes
-crew	adult	male	no
-third	child	female	no
-crew	adult	male	no
-second	adult	male	no
-second	adult	male	no
-second	adult	female	yes
-first	adult	male	no
-crew	adult	female	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-second	adult	male	no
-crew	adult	male	yes
-first	adult	male	no
-third	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	female	yes
-first	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-first	adult	female	yes
-third	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-first	adult	male	no
-second	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-second	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-crew	adult	male	yes
-third	adult	female	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-third	adult	female	no
-second	adult	male	no
-third	adult	female	yes
-third	adult	male	yes
-second	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-third	adult	female	no
-first	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	yes
-third	adult	female	yes
-third	adult	male	no
-second	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-third	adult	female	yes
-third	adult	male	no
-first	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-third	adult	female	yes
-second	child	male	yes
-third	adult	male	no
-third	adult	male	yes
-first	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-first	child	male	yes
-second	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-first	adult	female	yes
-crew	adult	male	yes
-third	adult	male	no
-second	adult	female	yes
-third	adult	male	no
-second	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-third	adult	female	yes
-first	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	yes
-third	child	male	no
-crew	adult	male	no
-crew	adult	male	yes
-first	adult	male	yes
-third	adult	male	yes
-first	adult	female	yes
-second	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-second	adult	female	yes
-crew	adult	male	no
-first	adult	male	yes
-second	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-second	adult	female	yes
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-second	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-first	adult	male	yes
-first	adult	male	no
-crew	adult	male	yes
-second	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-crew	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	yes
-crew	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-second	adult	female	yes
-first	adult	male	no
-third	adult	male	no
-second	adult	male	no
-second	adult	male	yes
-first	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-first	child	female	yes
-second	adult	female	no
-first	adult	female	yes
-first	adult	male	no
-second	adult	male	no
-second	adult	female	yes
-second	child	female	yes
-third	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-first	adult	female	yes
-crew	adult	male	no
-first	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-first	adult	female	yes
-second	adult	female	yes
-first	adult	female	yes
-first	adult	female	yes
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-first	adult	female	yes
-third	adult	female	no
-first	adult	female	yes
-third	adult	female	yes
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	male	yes
-third	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-crew	adult	male	yes
-third	adult	female	no
-crew	adult	male	no
-third	adult	male	no
-second	adult	female	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-first	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-second	adult	male	no
-crew	adult	male	no
-first	adult	female	no
-third	adult	male	no
-first	adult	male	no
-third	adult	female	yes
-first	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-first	adult	male	no
-first	adult	male	no
-third	child	female	no
-crew	adult	male	yes
-third	adult	female	no
-crew	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	female	no
-third	adult	male	no
-crew	adult	male	yes
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-third	adult	female	yes
-first	adult	female	yes
-second	adult	male	no
-third	adult	male	yes
-third	adult	male	no
-crew	adult	female	yes
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-third	adult	female	yes
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-third	child	female	yes
-third	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-second	adult	female	yes
-third	adult	female	yes
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-second	child	male	yes
-first	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-first	adult	male	yes
-first	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-crew	adult	male	yes
-first	adult	female	yes
-third	child	female	no
-third	adult	male	yes
-second	adult	female	yes
-first	adult	male	no
-first	adult	male	no
-third	child	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	female	yes
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-second	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-third	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-first	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-second	adult	male	no
-third	child	male	no
-third	child	female	yes
-crew	adult	male	no
-third	child	female	no
-second	adult	female	yes
-first	adult	male	no
-second	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-third	child	male	no
-second	adult	male	no
-first	adult	female	yes
-third	child	female	no
-crew	adult	male	yes
-crew	adult	male	no
-second	adult	male	no
-third	adult	female	yes
-third	child	male	no
-third	adult	male	no
-second	adult	male	no
-third	adult	male	no
-first	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	yes
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-first	adult	male	yes
-second	child	female	yes
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-second	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-first	adult	female	yes
-crew	adult	male	no
-third	child	male	no
-crew	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-third	adult	female	yes
-third	adult	male	no
-crew	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-second	adult	male	no
-crew	adult	male	yes
-second	adult	female	yes
-third	adult	male	no
-third	adult	female	no
-third	adult	male	no
-third	adult	female	no
-third	adult	male	no
-second	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-third	child	male	no
-third	adult	male	no
-first	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-third	adult	male	no
-second	adult	female	yes
-first	adult	male	yes
-third	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-second	child	female	yes
-second	adult	male	no
-second	adult	female	yes
-crew	adult	male	yes
-third	adult	male	no
-second	adult	male	no
-first	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-second	child	male	yes
-second	adult	female	yes
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-crew	adult	male	yes
-first	adult	female	yes
-second	child	female	yes
-third	adult	male	no
-first	adult	female	no
-crew	adult	male	no
-second	adult	male	no
-first	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-third	adult	male	yes
-second	adult	male	no
-third	adult	male	no
-third	adult	male	no
-first	adult	male	yes
-second	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-second	adult	male	yes
-first	adult	female	yes
-crew	adult	male	no
-first	child	male	yes
-first	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-second	adult	female	no
-third	child	male	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-third	adult	female	yes
-third	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	female	yes
-crew	adult	male	no
-third	adult	female	yes
-third	adult	female	no
-crew	adult	male	yes
-third	adult	female	yes
-crew	adult	female	yes
-third	child	male	no
-third	adult	male	no
-second	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-third	adult	male	no
-second	adult	female	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-first	adult	female	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	child	male	yes
-crew	adult	male	no
-second	adult	male	no
-second	adult	male	no
-second	adult	female	no
-crew	adult	male	no
-first	adult	female	yes
-third	adult	male	no
-third	adult	female	yes
-third	adult	female	no
-first	adult	male	no
-third	adult	female	yes
-first	adult	male	yes
-second	adult	female	no
-crew	adult	male	yes
-second	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-second	adult	male	no
-third	adult	female	yes
-second	adult	male	no
-first	adult	female	yes
-first	adult	female	yes
-crew	adult	female	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	child	female	yes
-second	adult	male	no
-first	adult	female	yes
-third	adult	female	no
-first	adult	female	yes
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	yes
-third	adult	male	no
-second	adult	male	no
-crew	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	female	no
-third	adult	male	no
-second	adult	female	yes
-third	adult	male	no
-third	adult	female	no
-first	adult	female	yes
-first	adult	male	no
-third	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-third	adult	male	yes
-crew	adult	male	no
-first	adult	male	no
-third	child	male	no
-first	adult	female	yes
-third	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-third	child	female	yes
-second	adult	male	no
-first	adult	female	yes
-second	child	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-second	adult	female	yes
-third	adult	male	no
-second	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	child	male	yes
-third	adult	male	no
-third	adult	male	no
-first	adult	male	yes
-crew	adult	male	no
-third	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-third	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-third	adult	female	yes
-crew	adult	male	yes
-third	child	male	no
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-second	child	male	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	male	yes
-third	adult	female	no
-first	adult	female	yes
-third	child	female	yes
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	female	no
-first	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-crew	adult	female	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-first	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-crew	adult	male	yes
-third	adult	female	no
-second	adult	male	no
-crew	adult	male	yes
-second	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-third	child	male	no
-crew	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-third	child	female	yes
-second	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	female	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-first	adult	female	yes
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-first	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-first	adult	female	yes
-third	adult	male	no
-third	adult	male	no
-third	adult	male	no
-third	adult	female	yes
-crew	adult	male	no
-crew	adult	female	no
-second	adult	female	no
-second	adult	female	yes
-third	child	female	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	female	yes
-third	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-first	adult	male	no
-third	adult	male	no
-third	adult	female	no
-first	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-third	child	male	no
-third	adult	male	yes
-second	child	female	yes
-third	adult	female	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-crew	adult	male	yes
-crew	adult	male	yes
-first	adult	male	yes
-crew	adult	male	yes
-third	adult	male	no
-crew	adult	male	yes
-third	child	female	no
-third	adult	male	no
-third	adult	male	no
-third	adult	male	no
-second	child	female	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-first	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-third	child	female	no
-first	adult	female	yes
-second	adult	female	yes
-second	adult	female	yes
-third	child	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-third	child	male	yes
-second	child	male	yes
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-second	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	yes
-first	adult	male	no
-crew	adult	male	yes
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-third	adult	female	yes
-crew	adult	male	no
-second	adult	female	yes
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	female	no
-third	adult	male	no
-second	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-third	adult	male	no
-second	adult	male	yes
-first	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-second	adult	male	no
-first	adult	female	yes
-first	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	female	no
-third	child	female	yes
-crew	adult	male	no
-crew	adult	male	no
-second	adult	male	yes
-crew	adult	male	no
-second	adult	male	no
-third	adult	male	no
-third	adult	female	no
-crew	adult	male	no
-third	adult	male	no
-third	child	male	no
-second	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-crew	adult	female	yes
-first	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-second	adult	male	no
-third	adult	male	no
-third	adult	female	no
-third	adult	male	yes
-crew	adult	male	no
-first	adult	male	yes
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-third	child	male	no
-third	adult	female	no
-third	adult	male	yes
-first	adult	female	yes
-first	adult	male	no
-third	adult	female	no
-third	adult	male	no
-crew	adult	male	yes
-third	adult	female	no
-first	adult	male	no
-third	adult	male	no
-second	child	male	yes
-third	adult	male	no
-crew	adult	male	no
-first	adult	male	yes
-second	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	yes
-first	adult	male	yes
-third	adult	female	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	female	yes
-crew	adult	male	yes
-crew	adult	male	no
-second	adult	male	yes
-second	adult	female	yes
-crew	adult	male	yes
-second	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-first	adult	male	no
-third	child	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	child	male	yes
-crew	adult	male	no
-crew	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-crew	adult	female	yes
-second	adult	male	no
-crew	adult	male	yes
-third	child	male	yes
-first	adult	female	yes
-third	adult	female	yes
-third	adult	male	no
-crew	adult	male	no
-first	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-second	adult	male	no
-second	adult	female	yes
-crew	adult	male	no
-third	adult	male	no
-third	adult	male	no
-third	adult	female	yes
-third	adult	male	no
-third	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-third	child	female	no
-crew	adult	male	no
-first	adult	male	no
-third	adult	male	no
-third	adult	male	no
-second	adult	male	no
-crew	adult	male	no
-third	adult	male	yes
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-first	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-third	adult	male	no
-crew	adult	male	no
-crew	adult	male	no
-first	adult	male	yes
-crew	adult	male	yes
-third	child	female	yes
-third	adult	male	no
-third	adult	male	yes
+status	age	sex	survived
+discrete	discrete	discrete	discrete
+			class
+third	adult	male	no
+second	adult	female	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+third	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	female	yes
+crew	adult	male	yes
+first	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	yes
+first	adult	female	yes
+second	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+first	adult	male	yes
+third	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+crew	adult	male	yes
+first	adult	female	yes
+third	adult	female	yes
+crew	adult	male	no
+third	adult	male	yes
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+third	adult	female	no
+crew	adult	male	yes
+first	adult	male	no
+first	adult	female	yes
+first	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	female	no
+second	adult	male	no
+second	adult	female	yes
+third	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+first	adult	female	yes
+third	adult	female	no
+crew	adult	male	yes
+crew	adult	male	no
+second	adult	female	yes
+first	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	female	yes
+crew	adult	male	no
+third	child	female	no
+crew	adult	male	no
+second	adult	male	no
+second	adult	male	no
+second	adult	female	yes
+first	adult	male	no
+crew	adult	female	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+second	adult	male	no
+crew	adult	male	yes
+first	adult	male	no
+third	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	female	yes
+first	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+first	adult	female	yes
+third	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+first	adult	male	no
+second	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+second	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+crew	adult	male	yes
+third	adult	female	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+third	adult	female	no
+second	adult	male	no
+third	adult	female	yes
+third	adult	male	yes
+second	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+third	adult	female	no
+first	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	yes
+third	adult	female	yes
+third	adult	male	no
+second	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+third	adult	female	yes
+third	adult	male	no
+first	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+third	adult	female	yes
+second	child	male	yes
+third	adult	male	no
+third	adult	male	yes
+first	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+first	child	male	yes
+second	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+first	adult	female	yes
+crew	adult	male	yes
+third	adult	male	no
+second	adult	female	yes
+third	adult	male	no
+second	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+third	adult	female	yes
+first	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	yes
+third	child	male	no
+crew	adult	male	no
+crew	adult	male	yes
+first	adult	male	yes
+third	adult	male	yes
+first	adult	female	yes
+second	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+second	adult	female	yes
+crew	adult	male	no
+first	adult	male	yes
+second	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+second	adult	female	yes
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+second	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+first	adult	male	yes
+first	adult	male	no
+crew	adult	male	yes
+second	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+crew	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	yes
+crew	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+second	adult	female	yes
+first	adult	male	no
+third	adult	male	no
+second	adult	male	no
+second	adult	male	yes
+first	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+first	child	female	yes
+second	adult	female	no
+first	adult	female	yes
+first	adult	male	no
+second	adult	male	no
+second	adult	female	yes
+second	child	female	yes
+third	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+first	adult	female	yes
+crew	adult	male	no
+first	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+first	adult	female	yes
+second	adult	female	yes
+first	adult	female	yes
+first	adult	female	yes
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+first	adult	female	yes
+third	adult	female	no
+first	adult	female	yes
+third	adult	female	yes
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	male	yes
+third	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+crew	adult	male	yes
+third	adult	female	no
+crew	adult	male	no
+third	adult	male	no
+second	adult	female	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+first	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+second	adult	male	no
+crew	adult	male	no
+first	adult	female	no
+third	adult	male	no
+first	adult	male	no
+third	adult	female	yes
+first	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+first	adult	male	no
+first	adult	male	no
+third	child	female	no
+crew	adult	male	yes
+third	adult	female	no
+crew	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	female	no
+third	adult	male	no
+crew	adult	male	yes
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+third	adult	female	yes
+first	adult	female	yes
+second	adult	male	no
+third	adult	male	yes
+third	adult	male	no
+crew	adult	female	yes
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+third	adult	female	yes
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+third	child	female	yes
+third	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+second	adult	female	yes
+third	adult	female	yes
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+second	child	male	yes
+first	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+first	adult	male	yes
+first	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+crew	adult	male	yes
+first	adult	female	yes
+third	child	female	no
+third	adult	male	yes
+second	adult	female	yes
+first	adult	male	no
+first	adult	male	no
+third	child	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	female	yes
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+second	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+third	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+first	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+second	adult	male	no
+third	child	male	no
+third	child	female	yes
+crew	adult	male	no
+third	child	female	no
+second	adult	female	yes
+first	adult	male	no
+second	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+third	child	male	no
+second	adult	male	no
+first	adult	female	yes
+third	child	female	no
+crew	adult	male	yes
+crew	adult	male	no
+second	adult	male	no
+third	adult	female	yes
+third	child	male	no
+third	adult	male	no
+second	adult	male	no
+third	adult	male	no
+first	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	yes
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+first	adult	male	yes
+second	child	female	yes
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+second	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+first	adult	female	yes
+crew	adult	male	no
+third	child	male	no
+crew	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+third	adult	female	yes
+third	adult	male	no
+crew	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+second	adult	male	no
+crew	adult	male	yes
+second	adult	female	yes
+third	adult	male	no
+third	adult	female	no
+third	adult	male	no
+third	adult	female	no
+third	adult	male	no
+second	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+third	child	male	no
+third	adult	male	no
+first	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+third	adult	male	no
+second	adult	female	yes
+first	adult	male	yes
+third	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+second	child	female	yes
+second	adult	male	no
+second	adult	female	yes
+crew	adult	male	yes
+third	adult	male	no
+second	adult	male	no
+first	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+second	child	male	yes
+second	adult	female	yes
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+crew	adult	male	yes
+first	adult	female	yes
+second	child	female	yes
+third	adult	male	no
+first	adult	female	no
+crew	adult	male	no
+second	adult	male	no
+first	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+third	adult	male	yes
+second	adult	male	no
+third	adult	male	no
+third	adult	male	no
+first	adult	male	yes
+second	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+second	adult	male	yes
+first	adult	female	yes
+crew	adult	male	no
+first	child	male	yes
+first	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+second	adult	female	no
+third	child	male	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+third	adult	female	yes
+third	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	female	yes
+crew	adult	male	no
+third	adult	female	yes
+third	adult	female	no
+crew	adult	male	yes
+third	adult	female	yes
+crew	adult	female	yes
+third	child	male	no
+third	adult	male	no
+second	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+third	adult	male	no
+second	adult	female	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+first	adult	female	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	child	male	yes
+crew	adult	male	no
+second	adult	male	no
+second	adult	male	no
+second	adult	female	no
+crew	adult	male	no
+first	adult	female	yes
+third	adult	male	no
+third	adult	female	yes
+third	adult	female	no
+first	adult	male	no
+third	adult	female	yes
+first	adult	male	yes
+second	adult	female	no
+crew	adult	male	yes
+second	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+second	adult	male	no
+third	adult	female	yes
+second	adult	male	no
+first	adult	female	yes
+first	adult	female	yes
+crew	adult	female	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	child	female	yes
+second	adult	male	no
+first	adult	female	yes
+third	adult	female	no
+first	adult	female	yes
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	yes
+third	adult	male	no
+second	adult	male	no
+crew	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	female	no
+third	adult	male	no
+second	adult	female	yes
+third	adult	male	no
+third	adult	female	no
+first	adult	female	yes
+first	adult	male	no
+third	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+third	adult	male	yes
+crew	adult	male	no
+first	adult	male	no
+third	child	male	no
+first	adult	female	yes
+third	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+third	child	female	yes
+second	adult	male	no
+first	adult	female	yes
+second	child	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+second	adult	female	yes
+third	adult	male	no
+second	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	child	male	yes
+third	adult	male	no
+third	adult	male	no
+first	adult	male	yes
+crew	adult	male	no
+third	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+third	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+third	adult	female	yes
+crew	adult	male	yes
+third	child	male	no
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+second	child	male	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	male	yes
+third	adult	female	no
+first	adult	female	yes
+third	child	female	yes
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	female	no
+first	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+crew	adult	female	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+first	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+crew	adult	male	yes
+third	adult	female	no
+second	adult	male	no
+crew	adult	male	yes
+second	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+third	child	male	no
+crew	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+third	child	female	yes
+second	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	female	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+first	adult	female	yes
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+first	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+first	adult	female	yes
+third	adult	male	no
+third	adult	male	no
+third	adult	male	no
+third	adult	female	yes
+crew	adult	male	no
+crew	adult	female	no
+second	adult	female	no
+second	adult	female	yes
+third	child	female	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	female	yes
+third	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+first	adult	male	no
+third	adult	male	no
+third	adult	female	no
+first	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+third	child	male	no
+third	adult	male	yes
+second	child	female	yes
+third	adult	female	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+crew	adult	male	yes
+crew	adult	male	yes
+first	adult	male	yes
+crew	adult	male	yes
+third	adult	male	no
+crew	adult	male	yes
+third	child	female	no
+third	adult	male	no
+third	adult	male	no
+third	adult	male	no
+second	child	female	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+first	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+third	child	female	no
+first	adult	female	yes
+second	adult	female	yes
+second	adult	female	yes
+third	child	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+third	child	male	yes
+second	child	male	yes
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+second	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	yes
+first	adult	male	no
+crew	adult	male	yes
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+third	adult	female	yes
+crew	adult	male	no
+second	adult	female	yes
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	female	no
+third	adult	male	no
+second	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+third	adult	male	no
+second	adult	male	yes
+first	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+second	adult	male	no
+first	adult	female	yes
+first	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	female	no
+third	child	female	yes
+crew	adult	male	no
+crew	adult	male	no
+second	adult	male	yes
+crew	adult	male	no
+second	adult	male	no
+third	adult	male	no
+third	adult	female	no
+crew	adult	male	no
+third	adult	male	no
+third	child	male	no
+second	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+crew	adult	female	yes
+first	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+second	adult	male	no
+third	adult	male	no
+third	adult	female	no
+third	adult	male	yes
+crew	adult	male	no
+first	adult	male	yes
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+third	child	male	no
+third	adult	female	no
+third	adult	male	yes
+first	adult	female	yes
+first	adult	male	no
+third	adult	female	no
+third	adult	male	no
+crew	adult	male	yes
+third	adult	female	no
+first	adult	male	no
+third	adult	male	no
+second	child	male	yes
+third	adult	male	no
+crew	adult	male	no
+first	adult	male	yes
+second	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	yes
+first	adult	male	yes
+third	adult	female	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	female	yes
+crew	adult	male	yes
+crew	adult	male	no
+second	adult	male	yes
+second	adult	female	yes
+crew	adult	male	yes
+second	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+first	adult	male	no
+third	child	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	child	male	yes
+crew	adult	male	no
+crew	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+crew	adult	female	yes
+second	adult	male	no
+crew	adult	male	yes
+third	child	male	yes
+first	adult	female	yes
+third	adult	female	yes
+third	adult	male	no
+crew	adult	male	no
+first	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+second	adult	male	no
+second	adult	female	yes
+crew	adult	male	no
+third	adult	male	no
+third	adult	male	no
+third	adult	female	yes
+third	adult	male	no
+third	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+third	child	female	no
+crew	adult	male	no
+first	adult	male	no
+third	adult	male	no
+third	adult	male	no
+second	adult	male	no
+crew	adult	male	no
+third	adult	male	yes
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+first	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+third	adult	male	no
+crew	adult	male	no
+crew	adult	male	no
+first	adult	male	yes
+crew	adult	male	yes
+third	child	female	yes
+third	adult	male	no
+third	adult	male	yes
diff --git a/notebooks/slike/type12_error.jpeg b/notebooks/slike/type12_error.jpeg
deleted file mode 100644
index f46609934bcdd2997062fecb18dbde76d747ed71..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 9628
zcmbt)bx<5#)9>Q$zAO%bV8PwpeQ|eKBsgSoch_tPZi_>32?2sbu;9TVg1ZIyc%Jv(
zx4ydn-age+J>93er>4%C?q8pIUV7dH;HxMoD*zA>004xS0eD^oL;^5S(TE5MN&aEd
zmq7+3Cnq5z0|J5MR3zl&WF+Kd)Ko0A<Wx*lq@?uRbWE?=I5|1VX?TTrIRsfaIN1;|
zF)?v*aDn*vKsF{?CXWA?*#1@g=lZ{Fod1=d`v3&!fPMrdBm{Z@A^`#t0mAbj00aOa
zA|bpG;J<{5jDm)Qh=BfLRmKM(A|N86AR}X<p`xI?OnJEj5FiuMp`db0YZK91dGLg#
zHKGwS@XF|T7Bzj*wGL08nLQ;Tm5u1%^2#V~K0_yCl(Ttp1w{Ba(0`-62uKKs$S9~U
zw$}sz#20#bG-RazGXVjS0Ev!}8=2?dD8wE`Ur-pro>u_aNH2Z}kO%;hfaS{Kc<D@*
z;&=etuyb3hRd5c;ak}rv12$I-52|>20p!YjnP@dd=iyJJCumqUu>LYqm0{;MhVweE
z!d$odN8ET4uRYX|-E6QN(`fCdq`)Xp8KDouvg%!EXrD4bCWOZ`el0rDgwcGMtvvPy
z2Kt-t_ZeGiklQl`eF>2YX;+7Oblm!uy%b|tlG}s7R=l-J>*4{<*mYojtzAL9rhRAT
zuJ=G=d!!)bGbq7i(5fqZWJ5I*^LCHI)^lXmRwS(25iSIhjpu{2aSgfz!MT3PZksCz
z)%<CX#IDba#yK`bB1b(e%ovb`=xY2v{Hrq_n?b@DnFmO8=nfENX%?xNxs&?qx>qZY
z6gw$TtVfK=QXKk!S&Sl+l9cR%<T$hS_^@<U7E5?wXslY$c~9rAi8dDYN2#%K&slju
z`4h6^k-#fU-<Yq{3VT2$`UHI5^g)1GY!~61RDW=Lj<(v;!I!Lnv?G!NmqgS>H3iBY
zLZy)^TrWxmT{$bBF`$$(V?>^I9IauJ`-Q&%ag()gMjdi{0b-iMGk|U*)BL9awm8cb
zZQ!@?wQZ0(fgzQAP^oKe^H~W}L<(jKnLV*Ou|0n(?nvdBTzvF5`vwoF*F3p;<8=m_
zeGVb$TLcn&Wl7U0-p~iuMzG5`^M<8+)u7OOev<dD;#n=S+sGWk6x?()ZKgHRwC#5C
zlX-TyoTg#lbl-(6liLupBu*Wo)-u>zYn1=|SUS%y$1gX=Qe0T6El&*iUyQJ?h>0Nv
zV7$XLj(OKL{3$96nR{#ksU<S;$NOKcz2J_H=`{#sjy;wf-ydK4T+BI5(3vO9mVKg_
zX#zvnvH4nIC1H7lxqyoAQ~ju&bmV$0)p@wfAo&P#w8oe)c2dP#+<6>GE0+K1oIK;q
zsCS%~@R%DJr{zM+2>q@B$qz9u^bQkN5<1m%46wY()hz$T!e4c7%HD*SZ1m;Obe<f0
zj!ZjNj^LaNac6Ihl+WvGl02IU`>28_fN?($NU_oUz6j)}b|Z8V6W+<z*=$DvJ`UF5
z>rCv~yOXZ(y;^-}MEKsNz>e<A!@sLTOdJFqw>1+JQ-8u)7`RCZ;U!dc5bR))Myup5
zLK`rdDSXg$>P|(mKFVM)45yTSFNM(#5@lxov&Vmd(XFkfHRwf`(_~Z0vd3G<s))$(
z(BC6)WHpMC*^+H{!YxQ-@o7RLP0%&K0pd~L;KDCUZr~DoH)3s=jg^IRhX6gD58LsQ
zE?$pkfoHv})^Nmlmcke2Bji->ZcWpZ`AXR;9O`xM!KDCU%~#E^n#S>Id6xjSv<DD^
z!z1Yon}q|wC8LlVC@KHOLk}feX=KyKc@{Dh9YrGw<=$4G>C1*X^-e?2nhJ5)>`6X!
zmK$x6?am2e2>xU|!;8>n`B5=maN;*Kfm%JQCSdtfWm3<rR7XMs8b<^}_?L(*R`e63
z<<ZLf^76ySBx3AIYOD8;YlD?t_7FUxT7?W>u3gSVOh(5<@*gt3d9FK7rj$wUTMDr=
zSAH<uAi_g-CA*|C*eg$xT5Fk)h{f`~_@V^lzFPFc=dC8)Cm5=sE?-LM*QW9zgRtUS
zx;G5uF0T0Bev@dKHhN+lXY`pr`DMlY2o%BBYd=AiWPuK<&2hT5`guysfH<5=H6qA*
z+Lihovim9_OZQAj8!>>I(#r{v@6nS(*ZIY1_Q^7lNN*Lu-j;#d^S6YGtf77uIF<mr
z=8?W?*xw?Qg=c_4XCbb%9+q*%;USH~3b*AXW7N{=JsrOB&v5z_LX4l<#Z*L&^*L0I
z?eNr~;)EF~6k=^@1nJ@kDNM?_Pu!}phMTzx<y^`vHST%QIn;=~R*i;Teiil?N-=As
zrS6P^nUeCF&_#b6WbmZ6YW&P7H*djN%5lZ{BFz!TAiP|EECei>T|>&I$5Hsc8dssH
zHG!o*yk9XBFW=odgoE9+rz;`uu!$Qj#*_Eyy*uJT2&T@4si+-x+s;^So+7e=bobJ#
z**?-W!;|r!)#cbdHX6p1JLeRPD8Y%%Jz%<Xg_G+~<k&tfAw3mkp6RdSOWrX7jV^Ci
z3f5BYAjR&uoZA;P|A@Q;t^r0}Q{HD!9djB}E64~dG)6bGA7H;c`_otW*d=)E$&O5x
z3dUu(_aE7yM0b#nBbFoRg2;}}dTDdIBwsA#>&8<*sZ6*j35s|!dY)H-@oWQUpBA_I
zd8#Cq)2@lN<@LfrMjw^QL;)QxsOF9s+ilzGibFYyQcMuSH-Y)CSF8%?AJ?tMyB7a2
zSLX+;jKgVYsC!Z0Pv%;Y=||3W3xzG4ho}--yE1Hfa$E#@S14;o+C`>7gDcKwQRfmX
z-NYx}N^Za9ku(x=787HW6X_40)ZsYvHGQkB!`kH}sj@!!WPF)EVr0<bc7gCkgfKWI
z%Ae#9jfP~~Q+DNt_X%`6N!~&0AQh{(C$U_7{oHw3F5hx~b4~a(Ec`%#=jZp5zqaPh
zB<Q!bjr)PJKZ4<x1mMUtTuUoUI&yY-&o!oZYAJnAKi|2Ug=Q5KDmKP7HcnX$xHD5x
zs_|Uf*HOeHaNZeUH-1?C40~;X=C_*5wx^$zn~4c8>HbEkN{$2ZOEZjsI%(6ccO!s#
z?BXW0M`k+xQbnS*7DzkVN%y_$8;&_CUd7w;)Y~oFNj;8#rIe>&l^-1$^r#Y5^=`QM
zAe|xk3}9~jX3fWSKw%~nQP+lAJ5FXm@0beHM74BL;=X1fZ{EA~gb7(e!qMlAukaqe
zHFrrYwj{i=H;L+)C_d)d>kSQ8uR-YXGh2+b-2L+7K^Q#HFb2xGgvn##rB;&Hjtm$r
zjj|G+kC=*tBd67PmW?gZ>4AJe(j)}_H=a4~9>e2I1YNhd$@AVeR=y!r*bDblX>T7}
zyJUM<u`uqC&gu_$krsii48CQasutkvi>AOn<wfzgPGPaMVFRwb0-A4m_d+PYR|jrC
z=G>I=Vm|}cN=X0oJOixr$X5J{d*koS7-(P86~5k)kQZ;tye1TE_|;amVGK3LlNhS5
zzRppTtF-@713k{_DmQzqWMToM8#);~Ea9Xc-J$I*e-8TlI|&Noc6-$?n1VetOo8|`
zee->Y3<U%yS7Oxdt48ddvG_>apG~%qD`?;{9u_#$=2M&JUUWfRP=!^Z`i1qAdweJs
z*li({C%~XKthMCpAY<Y!&xQW4kR;WwOkb#pcMm7#G;-5=w9*jR`b-P#BUat*><%fC
zYA2on`z>Zab3tOUc>cK5ayr42{S2=si+XI&nRQA+;wYmMhOlsi;=2{t+wKKONhm2|
z#*{#8IYU!w-O&YEZ?K;f9fqrAZ@yd$@$gx0(9Ml(98b?h)Y?F6`Qn5L+Rcx_jSGYJ
zh-;4o#^y&yXxXBpL%WE9eEiLz6SXfeV8Zlx)01lm<P4al?Q1caBJkH$2k!94LHIxW
z^E2+w4lZ8@L)nw7xX+?X6550J8EUuH@Oa1KERFGoz(>+5Q6sna7Ab5`a=N9iUsrgK
zHl-v4Ax<I^p+R<H#gZn&Qw68@1N)z|=uh~tHQ4=<ZIfbXs%fyMM1t$<YAAzFZ~X^d
z>W&$ZYpm3}x-oj{4@%Fq-ImKeJmZU(ID+KqvU;mcSIc7m{<gsA@ef!T8UxHL7hO><
z9jN(Zq(yCy$v>r708Kq#ue^L4IdTZLr%!#(dwUxP^=%@5s0EJ3(@lt145fLEb2@Ou
z;}K=Glz`=hRo3r1^ut^RYg-_=Ae64*<(o9K>*(nO+`LkkLL9*S9fF~KZ$zbQvuIK$
z5$^uBE-6#8FFvPEsLI_udY-XZ4SFgs;Uc@xL4Tz$>CyI{GK2M5?Wv#>V;0u6#!vkI
z-3EJP^A*!y-aUcq7LmfXzDXoD2{C29l8#B@=J!e~hKC0m)~-8Ke**)nrax2%Jq_d}
zt}K_@_^DD)Z<Uy;S@tEZAMH&R>>w<f;14t-xJfaXxfUHs>`k3pt7>$3U?k~a&k{Hp
z;nH;U*)^O02v`t|8BTYtv>GrM`1Nifs(tmCDpIJJoQhzCj0%a`W~lD&O~_``nXA+L
z>?0|zB_*rC_?aP+@DWo}54!h*8!0$%oEsD-wqlBj6RtYUsr)_QJ+b&>%ThCF2=_zv
zj>9`cJoh=R*5{OwA6{ya+E-G=@hnIJw~n4%<I|HQ&j4Q?w%7ezUD9tk{)!#=Y#haU
z<eJJse#3|9{hP=Ht`btr>_i#x()5}LYxt~eT+v9j#eR+R6MF|53Ukm1*dyCEJFZnn
zvtbjjTn;qP3p3&ZgZkk&;jb`fSzfJW4;1JQj~Iw|Whd8|Ddb6RK0yAOkX8<Nh?hJz
zL1R{wt`>TDlYQd8qB*=Gm3~{4m0Z5njuVK{<4hhn)$#sbe}breCSp8F<!jGo0{d`L
znR`-=dRiQ|Mrg8xVwC6!v9pRGCu+N^cr;T&Cm9#jtU^hCm|zR907?4mhazeTzj$>h
zHx^a0uXW?#?uEeC!Jav?8XW1JsmVYp^~FIoyUQzKT(F1-DgDYpki5*GHOkfaem5Hq
zMCpLJr)k)CYcf^`t(c-GUgmNCQ;h^ZJqjphV=7ya+1`#K_C&b1J1vOdNU%apW?8-3
z7k%DkKY(aYoTycDqQ3qNn5T6ub;tBrKj>^{99-^YVI@xtR-%`^NVy)?oT*4s8LpjM
zHl;Pc{1C*wc<H3mDBB+Rxz76IJ^#SccrX5Sj`fy8v{Gt;q849UAs)eE4y)`6Oje34
zqLT6Av$#{i78EbQ>aBmj^p&VdUTIHw%`u@WK5pm*+4Q^il?Dg><+;AgG)mw}<}O7H
zp^F>efm#9T>e#8Z0ua)q2A1B`zN7R)y)Nh2>thtMRAn`F<fx|fE}cvZ#fuB&rn<`v
zS4IZ(r4{bH(+B0isWNbDdHwqd&1%^F&0^?=2J*j20CD;80tB3L@aT(#$JV$-(#4#|
zkT$Lilu0*}3*#4AnBT2(DIR4Ne6Z@1aL}hZSR*>7QYi4zclE(I0IaT~h`X;(W#F;d
zQT6i)YzyBX2!0h+QE)T17`oxl!5Pz^ynC%^{xfFKj%{IOHR(3uXMeAiL*s;GonO6B
ztc%{qz>t*AVs2K5b$mKcN4uz`Od#$F##hGJfga4l$*aul@eqQ{g*3y@PaoBHhaPjq
z&lhIob(Hq+IL}*O3K-P4#ql@*kCdM@hZXk+oU1V5hNT##C%3IE^+!~}>15JIa~g$D
zGb~0g_MYF8L1##<kx$il-6j`h0Upl)vwTfNb@exyoC=8M{e7@S?7)DWPNhG>1~umV
zUclY<7BT+QI;ZcN)=$%`(qzqtD~gL!-c-2Z+bu~+i4Fsbd6=Z~jMTJQWN1T88;8^1
zvZp555<*>}A0v`tj2E}Q>>(%u0E<bX4RtXI;1ay|s}MUkL9CBrJviP2uK~DOj_u|x
z2aQ^s0s6hhB}gglC7)sBV+pM0XZKd}qlsy`q5RgZlWsYs%UvK&PNp38X|aF>rBT|n
zN=8W8oqznoguD^?MmMEK_%DTg{{pjWfm^L5l{uM);-W#GYCF;{`$ii#U#k$-Pjm@y
zew;lbH`|8qkS!f_+`clL`1X5D)uiyA*5A0E@x)}+jB6TwM%$k?idqUF{(XX@-mQ1M
z;P3akWw|OM0n0o73-_V#oW+OVVAQU+Nh+n^<|j@XisM;J@p)6vA#SJlpG!0<&Z~ue
zEWie=H@j^MR3QraXg~F76wM2=guYX!Wdw3qP5FcNN3APlExo*^ohq+&uJ;eCZ3s#)
z$;4yyNn8F3t&X)&by3@?BYeMmeNX$8^bBCNSjr8_Y8D71hZDe4$1md}Y=V&k^xx3G
z`Q#(!KrQJ(1oh?fx+(RY3q|e2&8}=vjxw}T0yhsS<5=@g*886=)Lmcev`#zv-+@s_
zm%_>u%XTphgc-A_<zRRK6sri=GZAHZ*$lpJUc$c9Ipa)?79vdOIy{{Xo($?@ZLK2}
z)?Sd#bB7QNXn@m?etb`6*?;TC8?n$oY~op4Pl@VH-<*E{WlVtLPM(yVRN$FePLyFl
z);4y8#yBS=jHc|7H)K%ceHrIFPVdu&&96@EU8)>X6@lr>+ui^l=G3}G;}TMNwsHu~
zy{`-nxh8+a24{i)^}=`M0<<mxkjS;xBe{)%RiPhBv$9)hM!l4XZoaIq*99E*0!<5*
zv>XekQ8cCpYg~@^#|SusT#cfaR?&!JkAzrnw7ktLrV9LE=;<9s2&@F^5%_QX=2x;j
zDDiNEKx2Uze^0b`Ek7ETzAq`;b{LK`&1kA1i)?S-(YyU1%JO^Hsq(PACEMg2P}gfr
z6_n)^vqlD`u+_eAWnQiSX=skI1A!@1TI3~toiS_J9gCx{#MXnj`T!5q2#(5;(k@77
z-HL#|V`8Gbf3GecDhC7apSGUft=^|l_gD-Fx4oKb7yp!u>xy0tq9BAN={W~#GF^E{
z)U{|o&RV*!rrbH){mwy&HcX{IMOvLNAoQu}Rm?&GMzoF-2u$VYJz9&yHTP6u5KP9e
zmT~tSudiZ{C3*jp=T}7XR3g;KT4)E$nVi6JdZPW)UeOX@$M06%zxs89maWfRfa!O8
zIK5f8e)BhetY^GwjV(>!I?Mv!yua#fW*hev4-F^ioqlVL(ik3BpuhfZt9)V?@BT0>
zSfC0_WQUh-A@J9}Y}VnAz2B0yJc$nat$$d6&^w!oUvsq@;}!5CEBcc$FTqlju(|y?
z<6P8&#oiB2F-m2FE(^IVj4TG9{drC88WE3%C*D=5Qe<Ph`~k?_3C$+~+$NN&_m@ir
zPb=2cmW@ORgT8Af^p)mTZKb(`&g6zC98CMp=|^js8HK-;n6oDnHVV3`t_cFy`{7(&
zRt?QxAikb~u!yU;&Vud~JJoe27xoA|A?v~n(0ilBqDQL<onu~5`Z#9crCm}BLq^2j
zScBED(AayC0MCBIBVN<x++9x>mob&jHWI5rnlAH5U=wQQO|QVfwk~n_Ob5mrsWtl4
zTS)51sCVEYGzuSVA!^06-zHOjl==yo{4`WiQJHz#=zpYc7Ui5Hx^-4i7j^U?KcBwa
zW*ma2^8|J*Z3pA%6W5o+E!vg$4F-~ow*=K0tXCkz!}XO0u6b?7?0m=LZ)34-6Oigm
z%R2tJ@9A|_w0PVu!sHyIM%a0xAPI6Yxea1I#4X|Zt^9Ym?M0t0t+Iv9z+u9~slwQk
z?jG0RN3hEWG~4~3yzLW5J5UrFIBLZB;P+Z{MbS64QN-F>EG&Flj))O1cv+HGK__}%
zinUxvOx+cS%sqc9h>A#R-)a5o!7rq+13&tYT(wt>sei;$#==EeJga)*hxVA*Z2K!W
z3iDp8!Z^-1v^b!Y9(4wHLLDj?!HPETSc3_!`Pvg-2MUBF$;kHE(3x=48M!HuvZ{ai
z4JNP6O6kv~?Ta~NNXAF!S{hxoz+Xy#+{PqS(C|C05(;`qS4id?U=vjp5iqI?m8m>D
z>($IW{)~kq-|ftawZ$@uk!dA`Qiy46LQBN0Wfz!}!qGCF<jof=%V%S%TVF8hO5Gp?
z3V!9pbvvJF<r>?yA3qVylM?efGn+J#pps;m26YK@DqT(0x&no)VIuVf!;v{2qkLg<
z_ktn}7~ebtI@GgC$+h(56rt+B*sPy8ejNK<HSP^{?MM*FsK+H)8Q{yu`Pil2BvN>l
zVCRo#GpMbxet-Xnb$yM4Yr@cSNPjfKvui#biWb5$S;&a<4tUmPgA<dbi5N-4P*mEw
z@j<p5vq7FhXf7sGtc~;#c|7YpNy+Vl{@O99_0na@35<?CB9qZaRQRfS7*zD0)BNNC
z?b8_5`tW`@p+<+41QWFM@w-!eDy071d~K5e@pSA?0H5?AE_2>uRgbZCYhY9}?+-M(
zqKE2f73D^4y>tp+RV2c+zKt_rj@Gr}L~VBnK~-kleO~^5kQFxKg(RL@ki`tMSx|U|
zCLM59Bbb{18!B`*sPP%#%2v{n+>M7PVw+b@BZpH3uLQZd?xP&(!M?WHL^U*Lt2X$s
ze#_qBgxsO3T2zDg=LFO?w4$#)9;#$<Qms8$iMy^_k71ymbe`9Oq|)yd+OS~X)nA)o
zQ|j7(ZQ+t4>^aM$=AmRbw{r`^GlP%}Z?0JbnNymxOi{tV(8mt8GLdesYTa$d&s+@b
z^#_Qf!fB5&7)Whkq4uVoK03~yq1UE{+pi)?2K#}12@a&<{r;5H)WQ6wDf$uAWtv3=
z&Q^qu>l7%nO6P59CP0lJVT=l<jFaz`YVGfT`6R3Qht<dOP3P9klz5FGP~o(cxo)L?
zjT_j`-cuY=x2iON4DD}#FHf-@P2{s<K<sT~aE{Z#RItoGEs<N~g+!hCs|pupNtO*c
z{+NUi{VHvj83xi|vc<tmYP}IG*m~q+OsD5|zWv}v6Y7@o_t4G4Ad<7w+5G)HT5XVX
z^=2VJ3;;UlogQN6&clg{j#zswX*snN*QLN%X;ZcD=IWehLPjjKH5_75^#soY!z2RF
ziC%!Z#80a`aK4U$OK<r<-7CYFo&$nr%H7Vj?)sHci%m<Cp)2ZL$dV>>-W8scc&p!j
z8C@6a=B3b!Eg2QG5>Ots`F>Vzdy1Sg?QIYf?RjP6_}W!{x85Eko;jK4ir)H{<R>G{
zsAPPU^E3+5&y;?jK8u*IbeV04aUiSkqN4IL+NOPJW<Yw^dw7g_0b(t1-txj8n6>bE
zb9wu)ZIVH2)fTNL1|i%m90?yUk;rR)C%1M}=kv2xeHr?leHPwI2&lU=Iwyp#@Lw%v
z9VCNLiL;u{Vo*~2K{r$vbj)EkZ-d&X;lsQBf60Un7sKC<+>UCu^-^joPz&DG*+{MD
z&;W1fNiSEP0ESS0<^{-vf8hvl(-f)ddXYELqNaC7J(An0G_r+ZEp^R*5w^5+N}s$$
z?X$IeHTSDX)Gk2*E#W1N`4HLBq+6`~RoD5gd&SFp_;Ny3g&au#w>0J|<^|H!^&BDc
z5PG*sMBf{H;5FKNkbPP)z^P4;D#Msh`8CIQ=`z?UET2e7JsA~8DvlR>3jdK24gM&b
ziN=CNT_)Nl?iMYwiJjeTXBbv?GMq)ermss)n^;?{(zfXerO~2)!cuwZe$+__?Tq{p
z{WZQW@}aHK0->HY19<I4G*Zxh3QyYMxYS7Qe0K|M`FKPqDHf|)EB|y);4F3|bUtaf
zIiCAupL&g~yKO-hO{T}&@un_K=}TQFq{vdtY24MOSK**0F`5Y}&cP1pxJP+eRQ&Df
z@2a+-p{NnG&WdDhuloG<?<I|1-f>GVw0ed+$HiEtumUQ7Qn5+@p`GGHEPT&Xq?|Je
zzPwM|hUH#ayJN}?2r{Ak(;f}e=5TJEDZW?FfJLCQOrGSs>rc7ST6TJI-d=T)OO9yN
zv;iAM)&8Yw0u=F7Brak{RTrh1I))S&f+CJCI7cDW!E**gO2u3s6qByRG;7x8XE2hx
z3SQaD_~#NAm^79fYp>}JT1aEtoLC24{!krVhA~&Y<SjiGVjxQ)CQg$poj8g3iG@8y
z>TUd8Yp+Ld|HFF67GozKoBqhX)*z?Sg4*XAw1#Iui;-0+T)c>mJX=EwmD=_nh2`jM
zy_8W$BeH>*cZkwk{@I&)CEnM%f8g0<%;uBN0PyE(9f$s4q%KmX08nBeF8pOFc7_aZ
zO~$uX-d@4YN+QV<`BK|Rbr8gIo$)?l#wlhRqpK!n&>btB{<Y?u{5D2Q_GoqWIf<LJ
z8FS#;jYp{zkD$3|{`|8hf${Ol5lyljqxJ?+Bw5`pACL9o?o<_YCv~(vQU<g!$HNyg
zFN=l?k(U6r7q^0oYYTI7nr{u<2=|Dm^mc~S$zaU;IoF?w^g3gjf}Q?pmgT7wROQk9
zi^Un_$vId=ZiWi>%HHN6_wDNxw7$7Ht9MG&3qa@)9*W0!2H>z|Gg3%N`z-d<6@qtN
zzk0|2H5ZmX{JNK#5EbW-TY64y-)nR2okETU<g_lkpmME^4qZFA`Tl|2aTA&TkNP(r
zO`U-G54$T>vK^Q+oAv~(;}i(IoSrjbE<eWq82+BtTa)17&6ZdY^fBSm?K|7_K^&^P
zPLEcwz*JE$BlxO}uU1ohJ*TD3u^+^$uY4-Y^>x>-l?a1U4GY1iSGJzJ3RDkyaNfY*
zso+3kRassN<TFhEp0JPH$<RoZ$G}Fcx|F1u_2((jGtGxKDX2sSPlol@z6}4ITWI`L
zZe^sDtJk!DF!{ru(9O8Ov6}s7fL_?(M~TKmk_)G!QHF0TlJ;#B)Qres-y|Rhf8cQp
zxAOrzE9BW^g1x4Nm&ai`f1}VE9@%$mwVfZt4SHf(N^w0+G{SUE?q>JrV2EvW<%~hf
zREnQk^tNvt$YAY2cM_3<{>=k@f4uM-xk!+0PkdTnwF5Mm;AXBUOVugwXR`L3tYF`i
zxE+ivD;Hi$>p&N~V8&}QknilBbsE&!9*CsEwJ&3)E2`ADRs}HxaQp`Pz;}Hoyuh?C
zs50uG^EN<L52Zdi(<>)1XM&}Fn%^dBc~%=V`1E{RgTf6iOOiDLW(>lwFbi}Yh=Shg
zEX`@gB@SaJBTEs+PQ1y#YzO)4Lu1ztt+paRJOeVH0sVPbr7gySvdB!xCpJ?*hTuCq
z8Bf5+!|#x04@`gHxPi}gI4toM{S$Y`U-*fg)pe%Jn}|V^r<cToIfET9<B*clQFfg*
zt%N25R?}w>x#A{!21LI2E$>^(_$|%#4B&VMMDb)0?plP>=<t=Y+wR+^^R-J&lwLr%
zk#Q4FDqLPeA)tuAZBHbcP1lgJ3u&g6y||>?3@8pu;jX<2)GARJH+*bp&=YkL8H(ev
z>>pPcJWIwpP{BQ&j2y5K(+&H~ct%X13=MBhx=nf)Vo+wvd2HkEo}keyo?-IC(90%=
zEpF!Qd^ky|L;ccddz@>O2Uex!*?3dM@6>dY${#^}_1Z?zfb_>!bS~F#%B9=WI<7<P
zcPmS3ihtYYj<<9%ltt39QzHyW-C;Nv(iEUCl2^cduxMp+yAX0bco33HV2#HmZ=$C;
zKLz<WZy-+17Z9wDqz*GAR2|c{Jn-)@v}-@q>a?1pf={enN`E2tdJX)>ha>D1lERO=
zdK9tv(pb&n^n2=u_^D0dB`DoOaX%BekJH`WYzE-GPW+L4Thy_OHTXN2`J%FYawkdy
zoQCxln4CqgG+^dFeo@wKe~5keg}61ke$*274asq?_8YzuO}+!=_bAB~W)jekIiLw0
zG8H=Q(#is9<AGTtWqo|W5~Cl`uion5RZT-z>PYR1S{sCtmJx?z@N91^dp&wsQ?cFS
zfG7EnfE1tCK-PSkkAxpuE%eLIPIHAXr7*Ax0K>`{p@jLbT%n<LmNQS;Vk_jdgi)cl
z{d8`Ep);iLdB9;<by9i-1cNjdBBtCeZrFj%Ltu*fdxP!PasuO7HCx72RxCWH`THmX
zSB`GX$V7IfH+I^;5-#tDlm6h&erP_$$PI;pgqc%jS|e}Gt7+_Z81;Py+NCc`r_-QF
znP&A{&ZdpFqf5_#su<vd>aVtf<{3AXI!)2AVDH}&XBhev@?%QjPh>P{$iSgS(UwG4
zqY|Vd_L4=PYO0S*6Jk`kVfI?ex}eZd+)_hlpTz~IrEb3@r`(G>?QyuE`bsWu1DSwy
zy>9MlIdHrerL}wRgUUAw{Y1qa8PU?ue&&%huFm|$4$S{1Brxeq`33lwjpJDWFTH5S
ze|~deX(85<GER6gk-n(@Q6oSh$2I<OYPAeiv{;-bjvNyP?9@MpDPxRm+qd54bIj&d
zR(sklC^P9Blut4Il^ysKe{)83uX!)8UYsoVs2rF|<M2|A<KJiM842J#1HQUjzwE}2
zV))IW$|V1>R{jc2d5HCc5X3zCs5X?RQTtNK#G~hOp`N8mqZ2LLy%qcnuyL}yrp82h
zo8gDB!*(B@(p&edYIo#5{f;#HXW8gcY2%)4%UxP|&TlObNz_{hri6Bz%S`%$b;Qe`
z@}OnV-no|B)PZ6+r;aisCovJ7j!WsP7Rn>TQ2npsqSgo2m$x8|AWy9I|I294%l`vz
Ca^~j%

diff --git a/notebooks/src/107-1_nadzorovano_naivniBayes.ipynb b/notebooks/src/107-1_nadzorovano_naivniBayes.ipynb
index e7310e9..250f6ac 100644
--- a/notebooks/src/107-1_nadzorovano_naivniBayes.ipynb
+++ b/notebooks/src/107-1_nadzorovano_naivniBayes.ipynb
@@ -42,7 +42,7 @@
       "source": [
         "### Primer za ogrevanje\n",
         "\n",
-        "V letniku \u0161portne gimnazije imamo 20 u\u010dencev. Vsak od njih sodeluje pri enem od \u0161portov: ```kosarka```, ```nogomet```, ```gimnastika```. Njihovo vi\u0161ino smo ocenili \"na oko\" in vsakemu u\u010dencu pripisali eno od mo\u017enih vrednosti: ```nizek```, ```srednji``` ali ```visok```."
+        "V razredu na športni gimnaziji imamo 20 učencev. Vsak od njih sodeluje pri enem od športov: `kosarka`, `nogomet` ali `gimnastika`. Njihovo višino smo ocenili \"na oko\" in vsakemu učencu pripisali eno od možnih vrednosti: `nizek`, `srednji` ali `visok`."
       ]
     },
     {
@@ -55,7 +55,87 @@
       "source": [
         "### Warmup example\n",
         "\n",
-        "There are 20 pupils in a year of sports high school. Each of them participates in one of the sports: ```basketball```, ```football```, ```gymnastics```. We evaluated their height \"on the eye\" and assigned to each pupil one of the possible values: ```low```, ```average``` or ```high```."
+        "There are 20 pupils in one class of a sports high school. Each pupil participates in one of the following sports: `basketball`, `football`, or `gymnastics`.\n",
+        "We roughly estimated each pupil’s height by sight and assigned one of the following categories: `low`, `average`, or `high`."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 2,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>visina</th>\n",
+              "      <th>sport</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>visok</td>\n",
+              "      <td>kosarka</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>visok</td>\n",
+              "      <td>kosarka</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>visok</td>\n",
+              "      <td>kosarka</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>visok</td>\n",
+              "      <td>kosarka</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>srednji</td>\n",
+              "      <td>kosarka</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "    visina    sport\n",
+              "0    visok  kosarka\n",
+              "1    visok  kosarka\n",
+              "2    visok  kosarka\n",
+              "3    visok  kosarka\n",
+              "4  srednji  kosarka"
+            ]
+          },
+          "execution_count": 2,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "data = pd.read_table('../data/sportniki.tab')\n",
+        "data[:5]"
       ]
     },
     {
@@ -73,7 +153,9 @@
         ]
       },
       "source": [
-        "<font color=\"blue\"> Kako bi novemu u\u010dencu Marku, ki je ```srednje``` rasti predlagali najprimernej\u0161i \u0161port? </font>"
+        "Na šolo se je vpisal nov dijak, Mark. Naš cilj je mu priporočiti najverjetnejši šport glede na njegovo višinsko kategorijo.\n",
+        "\n",
+        "Za začetek poglejmo kako popularni so posamezni športi:"
       ]
     },
     {
@@ -84,44 +166,15 @@
         ]
       },
       "source": [
-        "<font color=\"blue\">How would you suggest the most appropriate sport for the new pupil Mark, who is of ```average``` height? </font>"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {},
-      "outputs": [],
-      "source": [
-        "data = pd.read_table('../data/sportniki.tab', skiprows=[1,2])"
+        "A new pupil, Mark, enrolled in the school. Our goal is to recommend the most likely sport for him based on his height category.\n",
+        "\n",
+        "To start, let's look at how popular each sport is:"
       ]
     },
     {
       "cell_type": "code",
       "execution_count": 3,
       "metadata": {},
-      "outputs": [
-        {
-          "data": {
-            "text/plain": [
-              "Index(['visina', 'sport'], dtype='object')"
-            ]
-          },
-          "execution_count": 3,
-          "metadata": {},
-          "output_type": "execute_result"
-        }
-      ],
-      "source": [
-        "data.columns"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": 4,
-      "metadata": {
-        "scrolled": true
-      },
       "outputs": [
         {
           "data": {
@@ -143,57 +196,69 @@
               "<table border=\"1\" class=\"dataframe\">\n",
               "  <thead>\n",
               "    <tr style=\"text-align: right;\">\n",
-              "      <th></th>\n",
               "      <th>visina</th>\n",
+              "      <th>nizek</th>\n",
+              "      <th>srednji</th>\n",
+              "      <th>visok</th>\n",
+              "      <th>All</th>\n",
+              "    </tr>\n",
+              "    <tr>\n",
               "      <th>sport</th>\n",
+              "      <th></th>\n",
+              "      <th></th>\n",
+              "      <th></th>\n",
+              "      <th></th>\n",
               "    </tr>\n",
               "  </thead>\n",
               "  <tbody>\n",
               "    <tr>\n",
-              "      <th>0</th>\n",
-              "      <td>visok</td>\n",
-              "      <td>kosarka</td>\n",
-              "    </tr>\n",
-              "    <tr>\n",
-              "      <th>1</th>\n",
-              "      <td>visok</td>\n",
-              "      <td>kosarka</td>\n",
+              "      <th>gimnastika</th>\n",
+              "      <td>3</td>\n",
+              "      <td>2</td>\n",
+              "      <td>0</td>\n",
+              "      <td>5</td>\n",
               "    </tr>\n",
               "    <tr>\n",
-              "      <th>2</th>\n",
-              "      <td>visok</td>\n",
-              "      <td>kosarka</td>\n",
+              "      <th>kosarka</th>\n",
+              "      <td>1</td>\n",
+              "      <td>2</td>\n",
+              "      <td>5</td>\n",
+              "      <td>8</td>\n",
               "    </tr>\n",
               "    <tr>\n",
-              "      <th>3</th>\n",
-              "      <td>visok</td>\n",
-              "      <td>kosarka</td>\n",
+              "      <th>nogomet</th>\n",
+              "      <td>2</td>\n",
+              "      <td>3</td>\n",
+              "      <td>2</td>\n",
+              "      <td>7</td>\n",
               "    </tr>\n",
               "    <tr>\n",
-              "      <th>4</th>\n",
-              "      <td>srednji</td>\n",
-              "      <td>kosarka</td>\n",
+              "      <th>All</th>\n",
+              "      <td>6</td>\n",
+              "      <td>7</td>\n",
+              "      <td>7</td>\n",
+              "      <td>20</td>\n",
               "    </tr>\n",
               "  </tbody>\n",
               "</table>\n",
               "</div>"
             ],
             "text/plain": [
-              "    visina    sport\n",
-              "0    visok  kosarka\n",
-              "1    visok  kosarka\n",
-              "2    visok  kosarka\n",
-              "3    visok  kosarka\n",
-              "4  srednji  kosarka"
+              "visina      nizek  srednji  visok  All\n",
+              "sport                                 \n",
+              "gimnastika      3        2      0    5\n",
+              "kosarka         1        2      5    8\n",
+              "nogomet         2        3      2    7\n",
+              "All             6        7      7   20"
             ]
           },
-          "execution_count": 4,
+          "execution_count": 3,
           "metadata": {},
           "output_type": "execute_result"
         }
       ],
       "source": [
-        "data[:5]"
+        "pd.crosstab(data['sport'], data['visina'], margins=True)"
       ]
     },
     {
@@ -204,7 +269,11 @@
         ]
       },
       "source": [
-        "Za za\u010detek poglejmo kako popularni so posamezni \u0161porti:"
+        "Splošnim verjetnostim razredov pravimo *apriorne* verjetnosti.\n",
+        "\n",
+        "Označimo jih s $P(Y)$, kjer je $Y$ spremenljivka razreda.\n",
+        "\n",
+        "V našem primeru $Y$ zavzame vrednosti {`kosarka`, `nogomet`, `gimnastika`}."
       ]
     },
     {
@@ -215,65 +284,88 @@
         ]
       },
       "source": [
-        "To start, let's look at how popular each sport is:"
+        "The general probabilities of the classes are called *a priori* probabilities.\n",
+        "\n",
+        "Let us label them with $P(Y)$, where $Y$ is a class variable.\n",
+        "\n",
+        "In our example, $Y$ takes on the values {`basketball`, `football`, `gymnastics`}."
       ]
     },
     {
       "cell_type": "code",
-      "execution_count": 5,
+      "execution_count": 4,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "sport\n",
+              "gimnastika    0.25\n",
+              "kosarka       0.40\n",
+              "nogomet       0.35\n",
+              "dtype: float64"
+            ]
+          },
+          "execution_count": 4,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "priors = data.groupby(\"sport\").size() / len(data)\n",
+        "priors"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "Najpopularnejši šport je `košarka`, s katerim se ukvarja 8 oz. 40% učencev. Naš prvi predlog je torej, naj se Mark ukvarja s košarko. \n",
+        "Vendar ta ocena ne upošteva višine učencev. Odraža le splošno priljubljenost posameznih športov v razredu.\n",
+        "\n",
+        "Kakšna je porazdelitev višin znotraj športa?"
+      ]
+    },
+    {
+      "cell_type": "markdown",
       "metadata": {
-        "scrolled": false
+        "tags": [
+          "en"
+        ]
       },
+      "source": [
+        "The most popular sport is `basketball`, with 8 or 40% of pupils participating in it. Our first suggestion is that Mark should play basketball. \n",
+        "However, this estimate does not take the pupil’s height into account. It simply reflects the overall popularity of each sport in the class.\n",
+        "\n",
+        "What is the distribution of heights in the sport?"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 5,
+      "metadata": {},
       "outputs": [
         {
-          "name": "stdout",
-          "output_type": "stream",
-          "text": [
-            "kosarka\n",
-            "    visina    sport\n",
-            "0    visok  kosarka\n",
-            "1    visok  kosarka\n",
-            "2    visok  kosarka\n",
-            "3    visok  kosarka\n",
-            "4  srednji  kosarka\n",
-            "5  srednji  kosarka\n",
-            "6    nizek  kosarka\n",
-            "7    visok  kosarka\n",
-            "\n",
-            "Sport (Y): kosarka, \u0161tevilo: 8, verjetnost P(Y): 0.400000\n",
-            "nogomet\n",
-            "     visina    sport\n",
-            "8   srednji  nogomet\n",
-            "9   srednji  nogomet\n",
-            "10  srednji  nogomet\n",
-            "11    visok  nogomet\n",
-            "12    visok  nogomet\n",
-            "13    nizek  nogomet\n",
-            "14    nizek  nogomet\n",
-            "\n",
-            "Sport (Y): nogomet, \u0161tevilo: 7, verjetnost P(Y): 0.350000\n",
-            "gimnastika\n",
-            "     visina       sport\n",
-            "15    nizek  gimnastika\n",
-            "16    nizek  gimnastika\n",
-            "17    nizek  gimnastika\n",
-            "18  srednji  gimnastika\n",
-            "19  srednji  gimnastika\n",
-            "\n",
-            "Sport (Y): gimnastika, \u0161tevilo: 5, verjetnost P(Y): 0.250000\n"
-          ]
+          "data": {
+            "text/plain": [
+              "visina\n",
+              "visok      5\n",
+              "srednji    2\n",
+              "nizek      1\n",
+              "Name: count, dtype: int64"
+            ]
+          },
+          "execution_count": 5,
+          "metadata": {},
+          "output_type": "execute_result"
         }
       ],
       "source": [
-        "for sport in pd.unique(data['sport']):\n",
-        "    subset = data.loc[data['sport'] == sport]\n",
-        "    \n",
-        "    print(sport)\n",
-        "    print(subset)\n",
-        "    print()\n",
-        "    \n",
-        "    py     = len(subset) / len(data)\n",
-        "    print(\"Sport (Y): %s, \u0161tevilo: %d, verjetnost P(Y): %f\" % (sport, len(subset), py))"
+        "data.loc[data['sport'] == 'kosarka', \"visina\"].value_counts()"
       ]
     },
     {
@@ -284,14 +376,11 @@
         ]
       },
       "source": [
-        "Najpopularnej\u0161i \u0161port je ko\u0161arka, s katerim se ukvarja 8 oz. 40% u\u010dencev. Na\u0161 prvi predlog je torej, naj se Marko ukvarja s ko\u0161arko. S tem rezultatom nismo najbolj zadovoljni, saj vidimo da med ko\u0161arka\u0161i ni veliko \u0161portnikov ```srednje``` vi\u0161ine. Razlog? Pri izra\u010dunu nismo upo\u0161tevali verjetnosti lastnosti oz. *atributa* o Markovi vi\u0161ini.\n",
+        "Vidimo da med košarkaši ni veliko športnikov `srednje` višine. \n",
         "\n",
-        "<div style=\"background-color:#00ccff; margin-left:50px; margin-right:50px\"> Splo\u0161nim verjetnostmi razredov, ki smo jih izra\u010dunali pravimo *apriorne* verjetnosti.\n",
+        "Pri prvem predlogu nismo upoštevali verjetnosti lastnosti oz. *atributa* o Markovi višini. Da bi za Marka, ki je `srednje` postave, pripravili bolj utemeljen predlog, moramo upoštevati tudi povezavo med višino in izbranim športom.\n",
         "\n",
-        "Ozna\u010dimo jih s $P(Y)$, kjer je $Y$ spremenljivka razreda.\n",
-        "</div>\n",
-        "\n",
-        "V na\u0161em primeru $Y$ zavzame vrednostmi {```kosarka```, ```nogomet```, ```gimnastika```}."
+        "Verjetnosti $P(X|Y)$ pravimo *pogojna verjetnost spremenljivke $X$ pri znanem $Y$*.  Opredeljuje verjetnost, da je v primerih razreda $Y$ atribut $X$ zavzame določeno vrednost. "
       ]
     },
     {
@@ -302,14 +391,11 @@
         ]
       },
       "source": [
-        "The most popular sport is basketball, with 8 or 40% of pupils participating in it. Our first suggestion is that Marko should play basketball. This result is not the most satisfying, as we see that among basketball players there are not many athletes of ```medium``` height. The reason? In calculating, we did not take into account the probability of the property or *attribute* about Mark's height.\n",
-        "\n",
-        "<div style=\"background-color:#00ccff; margin-left:50px; margin-right:50px\"> The general probabilities of the classes that we calculated are called *a priori* probabilities.\n",
+        "We see that among basketball players there are not many athletes of `medium` height. \n",
         "\n",
-        "Let us label them with $P(Y)$, where $Y$ is a class variable.\n",
-        "</div>\n",
+        "For our first suggestion we did not take into account the probability of the property or *attribute* about Mark's height. To make a more informed recommendation for Mark,  who is of `average` build, we should also consider how height is related to the chosen sport.\n",
         "\n",
-        "In our example, $Y$ takes on the values {```basketball```, ```football```, ```gymnastics```}."
+        "The probability $P(X|Y)$ is called *conditional probability of variable $X$ given $Y$*. It determines the probability that in the cases of the $Y$ class the attribute $X$ takes a certain value."
       ]
     },
     {
@@ -321,22 +407,9 @@
           "name": "stdout",
           "output_type": "stream",
           "text": [
-            "Sport (Y): kosarka, \u0161t. srednje visokih: 2, verjetnost P(X=srednji|Y=kosarka): 0.250000\n",
-            "    visina    sport\n",
-            "4  srednji  kosarka\n",
-            "5  srednji  kosarka\n",
-            "\n",
-            "Sport (Y): nogomet, \u0161t. srednje visokih: 3, verjetnost P(X=srednji|Y=nogomet): 0.428571\n",
-            "     visina    sport\n",
-            "8   srednji  nogomet\n",
-            "9   srednji  nogomet\n",
-            "10  srednji  nogomet\n",
-            "\n",
-            "Sport (Y): gimnastika, \u0161t. srednje visokih: 2, verjetnost P(X=srednji|Y=gimnastika): 0.400000\n",
-            "     visina       sport\n",
-            "18  srednji  gimnastika\n",
-            "19  srednji  gimnastika\n",
-            "\n"
+            "Sport (Y): kosarka, št. srednje visokih: 2, verjetnost P(X=srednji|Y=kosarka): 0.250000\n",
+            "Sport (Y): nogomet, št. srednje visokih: 3, verjetnost P(X=srednji|Y=nogomet): 0.428571\n",
+            "Sport (Y): gimnastika, št. srednje visokih: 2, verjetnost P(X=srednji|Y=gimnastika): 0.400000\n"
           ]
         }
       ],
@@ -346,9 +419,7 @@
         "    subset_x = subset_y.loc[data['visina'] == 'srednji']\n",
         "    p_xy = len(subset_x) / len(subset_y)\n",
         "    \n",
-        "    print(\"Sport (Y): %s, \u0161t. srednje visokih: %d, verjetnost P(X=srednji|Y=%s): %f\" % (sport, len(subset_x), sport, p_xy, ))\n",
-        "    print(subset_x)\n",
-        "    print()"
+        "    print(\"Sport (Y): %s, št. srednje visokih: %d, verjetnost P(X=srednji|Y=%s): %f\" % (sport, len(subset_x), sport, p_xy, ))"
       ]
     },
     {
@@ -359,16 +430,9 @@
         ]
       },
       "source": [
+        "Zanimivo! Verjetnost `srednje` višine je največja med nogometaši. \n",
         "\n",
-        "<br/>\n",
-        "Zanimivo! Verjetnost ```srednje``` vi\u0161ine je najve\u010dja med nogometa\u0161i. Ali podatek zado\u0161\u010da za spremembo prvotne odlo\u010ditve?\n",
-        "\n",
-        "<br/>\n",
-        "<div style=\"background-color:#00ccff; margin-left:50px; margin-right:50px\">\n",
-        "Verjetnosti $P(X|Y)$ pravimo <i>pogojna verjetnost spremenljivke $X$ pri znanem $Y$</i>.  Opredeljuje verjetnost, da je v primerih razreda $Y$ atribut $X$ zavzame dolo\u010deno vrednost. \n",
-        "</div>\n",
-        "\n",
-        "Katera verjetnost pa nas v resnici zanima? \u017delimo, da izra\u010dun upo\u0161teva Markovo vi\u0161ino in oceni verjetnost vsakega od \u0161portov. To je verjetnost\n",
+        "Katera verjetnost pa nas v resnici zanima? Želimo, da izračun upošteva Markovo višino in oceni verjetnost vsakega od športov. To je verjetnost\n",
         "\n",
         "$$ P(Y|X) $$\n",
         "\n",
@@ -376,7 +440,7 @@
         "\n",
         "$$ P(Y|X=srednji)$$\n",
         "\n",
-        "Za izra\u010dun te verjetnosti uporabimo"
+        "Za izračun te verjetnosti uporabimo..."
       ]
     },
     {
@@ -387,23 +451,18 @@
         ]
       },
       "source": [
-        "<br/>\n",
-        "Interesting! The likelihood of a `` medium '' height is the highest among footballers. Is the information sufficient to change the original decision?\n",
-        "\n",
-        "<br/>\n",
-        "<div style=\"background-color:#00ccff; margin-left:50px; margin-right:50px\">\n",
-        "The probabilities of $P(X|Y)$ are called <i>pogojna verjetnost spremenljivke $X$ pri znanem $Y$</i>. It determines the probability that in the cases of the $Y$ class the attribute $X$ takes a certain value.\n",
-        "</div>\n",
+        "Interesting! The likelihood of a `medium` height is the highest among footballers. \n",
         "\n",
-        "What probability does we really care about? We want the calculation to take Mark's height into account and evaluate the likelihood of each of the sports. That's the probability\n",
+        "Which probability are we actually interested in? We want our calculation to take Mark’s height into account and estimate the probability of each sport.\n",
+        "This is the probability \n",
         "\n",
         "$$ P(Y|X) $$\n",
         "\n",
-        "or. in Mark's case\n",
+        "or in Mark's case\n",
         "\n",
         "$$ P(Y|X=srednji)$$\n",
         "\n",
-        "We use this probability to calculate this probability"
+        "To calculate this probability we use..."
       ]
     },
     {
@@ -414,28 +473,29 @@
         ]
       },
       "source": [
-        "## Bayesov obrazec\n",
+        "## Bayesov izrek\n",
         "\n",
-        "Da bi izra\u010dunali verjetno razreda pri danih atributih $P(Y|X)$, potrebujemo verjetnost za vse mo\u017ene kombinacije razreda $Y$ in atributov $X$, ki jo ozna\u010dimo z $P(X, Y)$. Iz pravil o pogojni verjetnosti sledi:\n",
+        "Za izračun najbolj verjetnega razreda glede na določene atribute, torej $(P(Y\\mid X))$, potrebujemo verjetnosti za vse možne kombinacije razreda $Y$ in atributov $X$, označene kot $P(X, Y)$.  \n",
+        "Iz pravil pogojne verjetnosti lahko izpeljemo:\n",
         "\n",
-        "$$ P(X, Y) = P(X|Y) \\cdot P(Y) = P(Y|X) \\cdot P(X)$$ \n",
+        "$$ P(X, Y) = P(X \\mid Y) \\cdot P(Y) = P(Y \\mid X) \\cdot P(X) $$\n",
         "\n",
-        "<br/>\n",
-        "<div style=\"background-color:#00ccff; margin-left:50px; margin-right:50px\">\n",
-        "Iz \u010desar sledi <i>Bayesov obrazec</i> za izra\u010dun $P(Y|X)$:\n",
+        "Od tod sledi *Bayesov izrek* za izračun $P(Y \\mid X)$:\n",
         "\n",
-        "$$P(Y|X) = \\frac{P(X|Y) \\cdot P(Y)}{P(X)} $$ \n",
-        "</div>\n",
-        "<br/>\n",
+        "$$ P(Y \\mid X) = \\frac{P(X \\mid Y) \\cdot P(Y)}{P(X)} $$\n",
         "\n",
-        "Izra\u010dun verjetnosti razreda $Y$ pri znanih atributih $X$ je torej odvisen od apriorne verjetnosti razreda $P(Y)$, pogojne verjetnosti $P(X|Y)$ in apriorne verjetnosti atributov $P(X)$. <font color=\"blue\">V Markovem primeru torej:</font>\n",
+        "Izračun verjetnosti razreda $Y$ pri znanem atributu $X$ tako temelji na:  \n",
+        "- a priori verjetnosti razreda $P(Y)$,  \n",
+        "- pogojni verjetnosti $P(X \\mid Y)$ in  \n",
+        "- a priori verjetnosti atributa $P(X)$.  \n",
         "\n",
-        "$$P(Y|X=srednji) = \\frac{P(X=srednji|Y) \\cdot P(Y)}{P(X=srednji)} $$ \n",
+        "V Markovem primeru to pomeni:\n",
         "\n",
+        "$$\n",
+        "P(Y \\mid X=\\text{srednji}) = \\frac{P(X=\\text{srednji} \\mid Y) \\cdot P(Y)}{P(X=\\text{srednji})}\n",
+        "$$\n",
         "\n",
-        "<br/>\n",
-        "<br/>\n",
-        "\u010ce verjetnost ocenimo za vsako mo\u017eno vrednost razreda Y, torej {```kosarka```, ```nogomet```, ```gimnastika```}, dobimo odgovor na prvotno vpra\u0161anje."
+        "Če ocenimo to verjetnost za vsako možno vrednost razreda $Y$, tj. {`košarka`, `nogomet`, `gimnastika`}, dobimo odgovor na prvotno vprašanje.\n"
       ]
     },
     {
@@ -446,28 +506,29 @@
         ]
       },
       "source": [
-        "## Bayes form\n",
+        "## Bayes’ Theorem\n",
         "\n",
-        "In order to calculate the likely class for given attributes of $P(Y|X)$, we need probability for all possible combinations of the class $Y$ and attributes $X$, which is denoted by $P(X, Y)$. It follows from the rules on conditional probability:\n",
+        "To calculate the most likely class given certain attributes, i.e. $(P(Y\\mid X))$, we need the probability for all possible combinations of the class $Y$ and attributes $X$, denoted as $P(X, Y)$.  \n",
+        "From the rules of conditional probability, we can derive:\n",
         "\n",
-        "$$ P(X, Y) = P(X|Y) \\cdot P(Y) = P(Y|X) \\cdot P(X)$$\n",
+        "$$ P(X, Y) = P(X \\mid Y) \\cdot P(Y) = P(Y \\mid X) \\cdot P(X) $$\n",
         "\n",
-        "<br/>\n",
-        "<div style=\"background-color:#00ccff; margin-left:50px; margin-right:50px\">\n",
-        "What follows is <i>Bayesov obrazec</i> for calculating $P(Y|X)$:\n",
+        "This leads to *Bayes’ theorem* for calculating $P(Y \\mid X)$:\n",
         "\n",
-        "$$P(Y|X) = \\frac{P(X|Y) \\cdot P(Y)}{P(X)} $$\n",
-        "</div>\n",
-        "<br/>\n",
+        "$$ P(Y \\mid X) = \\frac{P(X \\mid Y) \\cdot P(Y)}{P(X)} $$\n",
         "\n",
-        "The calculation of the probability of the class $Y$ in the known attribute of $X$ is therefore dependent on the a priori probability of the class $P(Y)$, the conditional probability of $P(X|Y)$, and the a priori probability of the attribute of $P(X)$. <font color=\"blue\">V In the example of Mark, then: </font>\n",
+        "The calculation of the probability of class $Y$ given the known attribute $X$ therefore depends on:  \n",
+        "- the prior probability of the class $P(Y)$,  \n",
+        "- the conditional probability $P(X \\mid Y)$, and  \n",
+        "- the prior probability of the attribute $P(X)$.  \n",
         "\n",
-        "$$P(Y|X=srednji) = \\frac{P(X=srednji|Y) \\cdot P(Y)}{P(X=srednji)} $$\n",
+        "In Mark’s example, this becomes:\n",
         "\n",
+        "$$\n",
+        "P(Y \\mid X=\\text{average}) = \\frac{P(X=\\text{average} \\mid Y) \\cdot P(Y)}{P(X=\\text{average})}\n",
+        "$$\n",
         "\n",
-        "<br/>\n",
-        "<br/>\n",
-        "If we estimate the probability for each possible value of the class Y, then {`` `basketball``,` `football```,` `` gymnastics```, we get the answer to the original question."
+        "By estimating this probability for each possible value of the class $Y$, i.e. {`basketball`, `football`, `gymnastics`}, we can determine the answer to the original question."
       ]
     },
     {
@@ -490,12 +551,11 @@
       "source": [
         "for sport in pd.unique(data['sport']):\n",
         "    \n",
-        "    subset_y  = data.loc[data['sport'] == sport]        # vsi sportniki danega sporta\n",
-        "    subset_x  = data.loc[data['visina'] == 'srednji']     # vsi srednje visoki ucenci\n",
+        "    subset_y  = data.loc[data['sport'] == sport] \n",
+        "    subset_x  = data.loc[data['visina'] == 'srednji']\n",
         "    \n",
-        "    subset_xy = subset_y.loc[data['visina'] == 'srednji'] # vsi srednje visoki ucenci v danem sportu\n",
+        "    subset_xy = subset_y.loc[data['visina'] == 'srednji'] \n",
         "    \n",
-        "    # Izracunamo verjetnosti\n",
         "    p_y  = len(subset_y)  / len(data)         \n",
         "    p_x  = len(subset_x)  / len(data)\n",
         "    p_xy = len(subset_xy) / len(subset_y)\n",
@@ -505,14 +565,6 @@
         "    print(\"Sport (Y): %s, napoved P(Y=%s | X=srednji): %f\" % (sport, sport, p_yx))"
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "\n",
-        " "
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -521,7 +573,9 @@
         ]
       },
       "source": [
-        "## Implementacija Naivnega Bayesovega klasifikatorja"
+        "Večinoma ne bomo imeli opravka samo z enim atributom. Poglejmo za primer podatke o potnikih ladje [Titanic](https://www.kaggle.com/c/titanic).\n",
+        "\n",
+        "Podatki so že razdeljeni na učno in testno množico. Za začetek naložimo učne podatke."
       ]
     },
     {
@@ -532,33 +586,30 @@
         ]
       },
       "source": [
-        "## Implementation of the Naive Bayes Classifier"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {
-        "tags": [
-          "sl"
-        ]
-      },
-      "source": [
-        "*Naivni Bayesov klasifikator* predpostavlja, da so atributi neodvisni med seboj, pri znanem razredu.\n",
+        "Usually, we won't be dealing with only one attribute. For example, let's use the data from the [Titanic](https://www.kaggle.com/c/titanic).\n",
         "\n",
-        "$$ P(Y|X_1, X_2, ..., X_p) = \\frac{P(Y) \\cdot P(X_1|Y) \\cdot P(X_2|Y) \\cdots P(X_p|Y)}{P(X)} $$"
+        "The data is already divided into a learning and test set. Let's start by loading the learning data."
       ]
     },
     {
-      "cell_type": "markdown",
-      "metadata": {
-        "tags": [
-          "en"
-        ]
-      },
+      "cell_type": "code",
+      "execution_count": 8,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "status\n",
+            "age\n",
+            "sex\n"
+          ]
+        }
+      ],
       "source": [
-        "The *Naive Bayes classifier* assumes that the attributes are independent of each other, with known class.\n",
-        "\n",
-        "$$ P(Y|X_1, X_2, ..., X_p) = \\frac{P(Y) \\cdot P(X_1|Y) \\cdot P(X_2|Y) \\cdots P(X_p|Y)}{P(X)} $$"
+        "data = pd.read_table('../data/titanic-training.tab', skiprows=[1,2])\n",
+        "for c in data.columns[:-1]:\n",
+        "    print(c)"
       ]
     },
     {
@@ -569,34 +620,24 @@
         ]
       },
       "source": [
-        "##### Vpra\u0161anje 5-2-1\n",
-        "Dopolni implementacijo naivnega Bayesovega klasifikatorja, ki je definiran v spodnjem odseku. Dopolniti je potrebno del kode, kjer izra\u010dunamo \n",
-        "* verjetnostne porazdelitev razredov $P(Y)$\n",
-        "* verjetnostne porazdelitve atributov pri znanem razredu $P(X|Y)$\n",
-        "\n",
-        "\n",
-        "\n",
-        "### Sklepanje o podatkih\n",
-        "\n",
-        "V primeru diskretnih atributov lahko obe porazdelitvi dobimo s *pre\u0161tevanjem*.\n",
-        "* $P(Y)$ *Kolikokrat se v podatkih pojavi razred $Y$?*\n",
-        "* $P(X|Y)$ *Kolikokrat se v podatkih, ki spadajo v razred $Y$, pojavi atribut $X$?*\n",
-        "\n",
+        "## Naivni Bayesov klasifikator na podatkih z več atributi\n",
         "\n",
-        "<font color=\"blue\"><b>Kaj pa $P(X)$?</b></font> Ta verjetnost je v\u010dasih te\u017eko izra\u010dunljiva, posebej pri visoko dimenzionalnih podatkih, saj ni nujno, da bodo v podatki prisotne vse kombinacije atributov. Na sre\u010do ta vrednost ne vpliva na izbiro najverjetnej\u0161ega razreda za posamezen primer!\n",
-        "\n",
-        "### Napovedovanje\n",
-        "\n",
-        "Za nov primer $X^* = (X_1^*, X_2^*, ..., X_p^*)$ med vsemi vrednostmi razreda $Y=y$, izberi tisto, ki maksimizira naslednji izraz:\n",
+        "*Naivni Bayesov klasifikator* predpostavlja, da so atributi med seboj neodvisni pri znanem razredu. To nam omogoča, da skupno verjetnost izrazimo kot produkt posameznih pogojnih verjetnosti:\n",
         "\n",
+        "$$\n",
+        "\\begin{aligned}\n",
+        "P(Y \\mid X_1, X_2, ..., X_p) &= \\frac{P(Y) \\cdot P(X_1 \\mid Y) \\cdot P(X_2 \\mid Y) \\cdots P(X_p \\mid Y)}{P(X)}\\\\\n",
+        "&=\\frac{P(Y)\\prod_{i=1}^{n}P(X_i \\mid Y)}{P(X)}\n",
+        "\\end{aligned}\n",
+        "$$\n",
         "\n",
-        "$$ \\text{arg max}_y \\ P(Y=y) \\cdot P(X_1^*|Y=y) \\cdot P(X_2^*|Y=y) \\cdots P(X_p^*|Y=y) $$\n",
+        "Verjetnost $P(X)$ je včasih težko izračunljiva, še posebej pri visoko dimenzionalnih podatkih, saj v podatkih pogosto niso prisotne vse kombinacije atributov. Na srečo ta vrednost ne vpliva na izbiro najverjetnejšega razreda za posamezen primer. Zato lahko imenovalec izpustimo:\n",
         "\n",
-        "### Log-transformacija\n",
-        "\n",
-        "Te\u017eava pri zgornjem pristopu je prakti\u010dne narave; mno\u017eenje velikega \u0161tevila verjetnosti hitro privede do zelo majhnih \u0161tevil, ki lahko prese\u017eejo strojno natan\u010dnost. Najenostavnej\u0161a re\u0161itev, ki privede do enake izbire razreda je naslednja \n",
+        "$$\n",
+        "P(Y \\mid X_1, X_2, ..., X_p) \\propto P(Y)\\prod_{i=1}^{n}P(X_i \\mid Y)\n",
+        "$$\n",
         "\n",
-        "$$ \\text{arg max}_y \\ \\text{log } P(Y=y) + \\text{log } P(X_1|Y=y) + \\text{log } P(X_2|Y=y) + ... + \\text{log } P(X_p|Y=y) $$"
+        "Zanima nas torej vrednost $Y$, ki maksimira zgornji izraz, torej tista, ki ima največjo verjetnost glede na podane atribute."
       ]
     },
     {
@@ -607,30 +648,122 @@
         ]
       },
       "source": [
-        "##### Vpra\u0161anje 5-2-1\n",
-        "Complete the implementation of the Naive Bayes classifier, which is defined in the lower section. It is necessary to complete the part of the code where we calculate\n",
-        "* probability distribution of classes $P(Y)$\n",
-        "* probability distribution of attributes in the known class $P(X|Y)$\n",
+        "## Naive Bayes Classifier on data with multiple attributes\n",
         "\n",
-        "### Conclusion on data\n",
+        "The *Naive Bayes classifier* assumes that the attributes are independent of each other, given the class. This allows the joint probability to be expressed as a product of individual conditional probabilities:\n",
         "\n",
-        "In the case of discrete attributes, both distributions can be obtained by *counting*.\n",
-        "* $P(Y)$ *How many times does the $Y$ class appear in the data?*\n",
-        "* $P(X|Y)$ *How many times does the $X$ attribute appear in the data that belong to the $Y$ class?*\n",
-        "\n",
-        "<font color=\"blue\"><b>What about $P(X)$?</b></font> This probability is sometimes difficult to calculate, especially for high-dimensional data, since it is not necessary that all combinations of attributes will be present in the data. Fortunately, this value does not affect the choice of the most likely grade for a particular case!\n",
-        "\n",
-        "### Predicting\n",
+        "$$\n",
+        "\\begin{aligned}\n",
+        "P(Y \\mid X_1, X_2, ..., X_p) &= \\frac{P(Y) \\cdot P(X_1 \\mid Y) \\cdot P(X_2 \\mid Y) \\cdots P(X_p \\mid Y)}{P(X)}\\\\\n",
+        "&=\\frac{P(Y)\\prod_{i=1}^{n}P(X_i \\mid Y)}{P(X)}\n",
+        "\\end{aligned}\n",
+        "$$\n",
         "\n",
-        "For a new example  $X^* = (X_1^*, X_2^*, ..., X_p^*)$ among all values of the $Y=y$ class, select one that maximizes the following expression:\n",
-        "\n",
-        "$$ \\text{arg max}_y \\ P(Y=y) \\cdot P(X_1^*|Y=y) \\cdot P(X_2^*|Y=y) \\cdots P(X_p^*|Y=y) $$\n",
-        "\n",
-        "### Log-transformation\n",
+        "The probability $P(X)$ is sometimes difficult to compute, especially for high-dimensional data, since not all combinations of attributes are necessarily present in the dataset. Fortunately, this value does not affect the choice of the most likely class for a given instance. Therefore, we can omit the denominator:\n",
         "\n",
-        "The problem with the above approach is rather practical; multiplying a large number of probabilities quickly leads to very small numbers that can exceed machine accuracy. The simplest solution that leads to the same class choice is the following\n",
+        "$$\n",
+        "P(Y \\mid X_1, X_2, ..., X_p) \\propto P(Y)\\prod_{i=1}^{n}P(X_i \\mid Y)\n",
+        "$$\n",
         "\n",
-        "$$ \\text{arg max}_y \\ \\text{log } P(Y=y) + \\text{log } P(X_1|Y=y) + \\text{log } P(X_2|Y=y) + ... + \\text{log } P(X_p|Y=y) $$"
+        "We are therefore interested in the value of $Y$ that maximizes the above expression, that is, the one with the highest probability given the observed attributes."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 9,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "array(['no', 'yes'], dtype=object)"
+            ]
+          },
+          "execution_count": 9,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "class_variable = data.columns[-1]\n",
+        "variables = data.columns[:-1]\n",
+        "class_values = pd.unique(data[class_variable])\n",
+        "class_values"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "V primeru diskretnih atributov lahko obe porazdelitvi dobimo s *preštevanjem*.\n",
+        "* $P(Y)$ *Kolikokrat se v podatkih pojavi razred $Y$?*\n",
+        "* $P(X|Y)$ *Kolikokrat se v podatkih, ki spadajo v razred $Y$, pojavi atribut $X$?*"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "In the case of discrete attributes, both distributions can be obtained by *counting*.\n",
+        "* $P(Y)$ *How many times does the $Y$ class appear in the data?*\n",
+        "* $P(X|Y)$ *How many times does the $X$ attribute appear in the data that belong to the $Y$ class?*"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 10,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "{'no': 0.6745454545454546,\n",
+              " ('status', 'third', 'no'): 0.3450134770889488,\n",
+              " ('status', 'second', 'no'): 0.12398921832884097,\n",
+              " ('status', 'crew', 'no'): 0.4568733153638814,\n",
+              " ('status', 'first', 'no'): 0.07412398921832884,\n",
+              " ('age', 'adult', 'no'): 0.9663072776280324,\n",
+              " ('age', 'child', 'no'): 0.03369272237196765,\n",
+              " ('sex', 'male', 'no'): 0.9110512129380054,\n",
+              " ('sex', 'female', 'no'): 0.0889487870619946,\n",
+              " 'yes': 0.32545454545454544,\n",
+              " ('status', 'third', 'yes'): 0.24581005586592178,\n",
+              " ('status', 'second', 'yes'): 0.17039106145251395,\n",
+              " ('status', 'crew', 'yes'): 0.29329608938547486,\n",
+              " ('status', 'first', 'yes'): 0.2905027932960894,\n",
+              " ('age', 'adult', 'yes'): 0.9162011173184358,\n",
+              " ('age', 'child', 'yes'): 0.08379888268156424,\n",
+              " ('sex', 'male', 'yes'): 0.48044692737430167,\n",
+              " ('sex', 'female', 'yes'): 0.5195530726256983}"
+            ]
+          },
+          "execution_count": 10,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "n = len(data)\n",
+        "P = dict()\n",
+        "for y in class_values:\n",
+        "    data_subset = data.loc[data[class_variable] == y]\n",
+        "    m = len(data_subset)\n",
+        "    P[y] = m/n\n",
+        "\n",
+        "    for variable in variables:\n",
+        "        for x in pd.unique(data[variable]):\n",
+        "            data_subset = data.loc[(data[variable] == x) & (data[class_variable] == y)]\n",
+        "            p = len(data_subset)\n",
+        "            P[variable, x, y] = (p/m)\n",
+        "P"
       ]
     },
     {
@@ -641,11 +774,11 @@
         ]
       },
       "source": [
-        "Pri implementaciji si pomagaj s podatki potnikov ladje <i><a href=\"https://www.kaggle.com/c/titanic\">Titanic</a></i>.\n",
+        "### Napovedovanje\n",
         "\n",
-        "Podatke so \u017ee razdeljeni na u\u010dno in testno mno\u017eico.\n",
+        "Za nov primer $X^* = (X_1^*, X_2^*, ..., X_p^*)$ med vsemi vrednostmi razreda $Y=y$, izberi tisto, ki maksimizira naslednji izraz:\n",
         "\n",
-        "Nalo\u017eimo u\u010dne podatke in izra\u010dunamo verjetnosti."
+        "$$ \\text{arg max}_y \\ P(Y=y) \\cdot P(X_1^*|Y=y) \\cdot P(X_2^*|Y=y) \\cdots P(X_p^*|Y=y) $$"
       ]
     },
     {
@@ -656,219 +789,756 @@
         ]
       },
       "source": [
-        "For the implementation help yourself with the passenger data from <i><a href=\"https://www.kaggle.com/c/titanic\">Titanic</a></i>.\n",
+        "### Predicting\n",
         "\n",
-        "The data is already divided into a learning and test set.\n",
+        "For a new example  $X^* = (X_1^*, X_2^*, ..., X_p^*)$ among all values of the $Y=y$ class, select one that maximizes the following expression:\n",
         "\n",
-        "Load the learning data and calculate probabilities."
+        "$$ \\text{arg max}_y \\ P(Y=y) \\cdot P(X_1^*|Y=y) \\cdot P(X_2^*|Y=y) \\cdots P(X_p^*|Y=y) $$"
       ]
     },
     {
       "cell_type": "code",
-      "execution_count": 8,
+      "execution_count": 11,
       "metadata": {},
       "outputs": [
         {
           "name": "stdout",
           "output_type": "stream",
           "text": [
-            "survived\n",
-            "['no' 'yes']\n"
+            "status    third\n",
+            "age       adult\n",
+            "sex        male\n",
+            "Name: 0, dtype: object\n",
+            "no:\t 0.205\n",
+            "yes:\t 0.035\n",
+            "Predicted class:  no\n",
+            "Actual class:  no\n"
           ]
-        },
+        }
+      ],
+      "source": [
+        "row = data.iloc[0]\n",
+        "print(row[:-1])\n",
+        "predictions = dict()\n",
+        "for y in class_values:\n",
+        "    p = P[y]\n",
+        "    for x in variables:\n",
+        "        p = p * P[x, row[x], y]\n",
+        "    predictions[y] = p\n",
+        "    print(\"%s:\\t %.3f\"% (y, p))\n",
+        "print(\"Predicted class: \", max(predictions, key=predictions.get))\n",
+        "print(\"Actual class: \", row.iloc[-1])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "### Log-transformacija\n",
+        "\n",
+        "Težava pri zgornjem pristopu je praktične narave; množenje velikega števila verjetnosti hitro privede do zelo majhnih števil, ki lahko presežejo strojno natančnost. "
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "### Log-transformation\n",
+        "\n",
+        "The problem with the above approach is rather practical; multiplying a large number of probabilities quickly leads to very small numbers that can exceed machine accuracy. "
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 12,
+      "metadata": {},
+      "outputs": [
         {
           "data": {
             "text/plain": [
-              "0.08379888268156424"
+              "True"
             ]
           },
-          "execution_count": 8,
+          "execution_count": 12,
           "metadata": {},
           "output_type": "execute_result"
         }
       ],
       "source": [
-        "#data = Table('podatki/titanic-training.tab')\n",
-        "data = pd.read_table('podatki/titanic-training.tab', skiprows=[1,2])\n",
-        "print(data.columns[-1])\n",
-        "print(pd.unique(data['survived']))\n",
+        "a = P['yes']\n",
+        "for _ in range(1000):\n",
+        "    a = a * P['yes']\n",
+        "a == 0"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "Najenostavnejša rešitev, ki privede do enake izbire razreda je naslednja \n",
         "\n",
-        "# P(X=child | Y = yes)\n",
-        "survived_child  = data.loc[(data['age'] == 'child') & (data['survived'] == 'yes')]  \n",
-        "all_survived  = data.loc[data['survived'] == 'yes']\n",
+        "$$ \\text{arg max}_y \\ \\text{log } P(Y=y) + \\text{log } P(X_1|Y=y) + \\text{log } P(X_2|Y=y) + ... + \\text{log } P(X_p|Y=y) $$"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "The simplest solution that leads to the same class choice is the following\n",
         "\n",
-        "p_xy = len(survived_child) / len(all_survived)\n",
-        "p_xy"
+        "$$ \\text{arg max}_y \\ \\text{log } P(Y=y) + \\text{log } P(X_1|Y=y) + \\text{log } P(X_2|Y=y) + ... + \\text{log } P(X_p|Y=y) $$"
       ]
     },
     {
       "cell_type": "code",
-      "execution_count": 9,
+      "execution_count": 13,
       "metadata": {},
       "outputs": [
         {
-          "data": {
-            "text/plain": [
-              "Index(['status', 'age', 'sex'], dtype='object')"
-            ]
-          },
-          "execution_count": 9,
-          "metadata": {},
-          "output_type": "execute_result"
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "no:\t -1.585\n",
+            "yes:\t -3.346\n",
+            "Predicted class:  no\n",
+            "Actual class:  no\n"
+          ]
         }
       ],
       "source": [
-        "data.columns[:-1]"
+        "row = data.iloc[0]\n",
+        "predictions = dict()\n",
+        "for y in class_values:\n",
+        "    p = np.log(P[y])\n",
+        "    for x in variables:\n",
+        "        p = p + np.log(P[x, row[x], y])\n",
+        "    predictions[y] = p\n",
+        "    print(\"%s:\\t %.3f\"% (y, p))\n",
+        "print(\"Predicted class: \", max(predictions, key=predictions.get))\n",
+        "print(\"Actual class: \", row.iloc[-1])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "## Ocenjevanje uspešnosti klasifikacije\n",
+        "\n",
+        "Uspešnosti klasifikatorja ne merimo na posameznih primerih, temveč na večjem številu primerov, saj šele tako dobimo zanesljivo oceno njegovega delovanja."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "## Evaluating Classification Performance\n",
+        "\n",
+        "We do not measure the performance of a classifier on individual cases, but rather on a larger number of them, since only then can we obtain a reliable estimate of how well it performs."
       ]
     },
     {
       "cell_type": "code",
-      "execution_count": 10,
+      "execution_count": 14,
       "metadata": {},
-      "outputs": [],
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Actual\tPredicted\n",
+            "no \t no\n",
+            "yes \t no\n",
+            "no \t no\n",
+            "no \t no\n",
+            "no \t no\n",
+            "no \t no\n",
+            "no \t no\n",
+            "no \t no\n",
+            "no \t yes\n",
+            "no \t no\n"
+          ]
+        }
+      ],
       "source": [
-        "class NaiveBayes:\n",
-        "    \"\"\"\n",
-        "    Naive Bayes classifier.\n",
-        "    \n",
-        "    :attribute self.probabilities\n",
-        "        Dictionary that stores\n",
-        "            - prior class probabilities P(Y)\n",
-        "            - attribute probabilities conditional on class P(X|Y)\n",
-        "    \n",
-        "    :attribute self.class_values\n",
-        "        All possible values of the class.\n",
-        "        \n",
-        "    :attribute self.variables\n",
-        "        Variables in the data. \n",
-        "    \n",
-        "    :attribute self.trained\n",
-        "        Set to True after fit is called.\n",
-        "    \"\"\"\n",
-        "    \n",
-        "    def __init__(self):\n",
-        "        self.trained       = False\n",
-        "        self.probabilities = dict()   \n",
-        "    \n",
-        "    \n",
-        "    def fit(self, data):\n",
-        "        \"\"\"\n",
-        "        Fit a NaiveBayes classifier.\n",
-        "        \n",
-        "        :param data\n",
-        "            Orange data Table.        \n",
-        "        \"\"\"\n",
-        "        class_variable      = data.domain.class_var    # class variable (Y) \n",
-        "        self.class_values   = class_variable.values    # possible class values\n",
-        "        self.variables      = data.domain.attributes    # all other variables (X)\n",
-        "        \n",
-        "        n = len(data) # number of all data points\n",
-        "        \n",
-        "        # Compute P(Y)\n",
-        "        for y in self.class_values:\n",
-        "\n",
-        "            # A not too smart guess (INCORRECT)\n",
-        "            self.probabilities[y] = 1/len(self.class_values)\n",
-        "            \n",
-        "            # <your code here>\n",
-        "            # Compute class probabilities and correctly fill\n",
-        "            #   probabilities[y] = ... \n",
-        "            # Select all examples (rows) with class = y\n",
-        "          \n",
-        "            # </your code here>\n",
-        "        \n",
-        "        # Compute P(X|Y)\n",
-        "        for y in self.class_values:\n",
-        "            \n",
-        "            # Select all examples (rows) with class = y\n",
-        "            filty = SameValue(class_variable, y)\n",
-        "            \n",
-        "            for variable in self.variables:\n",
-        "                for x in variable.values:\n",
-        "                    \n",
-        "                    # A not too smart guess (INCORRECT)\n",
-        "                    p = 1 / (len(self.variables) * len(variable.values) * len(self.class_values))\n",
-        "                    \n",
-        "                    # P(variable=x|Y=y)\n",
-        "                    self.probabilities[variable, x, y] = p\n",
-        "                    \n",
-        "                \n",
-        "                    # <your code here>\n",
-        "                    # Compute correct conditional class probability\n",
-        "                    #   probabilities[x, value, c] = ... \n",
-        "                    # \n",
-        "                    # Select all examples with class == y AND \n",
-        "                    # variable x == value\n",
-        "                    # Hint: use SameValue filter twice\n",
-        "            \n",
-        "                \n",
-        "                    # </your code here>\n",
-        "    \n",
-        "        self.trained = True\n",
-        "        \n",
-        "    \n",
-        "    def predict_instance(self, row):\n",
-        "        \"\"\"\n",
-        "        Predict a class value for one row.\n",
-        "        \n",
-        "        :param row\n",
-        "            Orange data Instance.\n",
-        "        :return \n",
-        "            Class prediction.\n",
-        "        \"\"\"\n",
-        "        curr_p = float(\"-inf\")   # Current highest \"probability\" (unnormalized)\n",
-        "        curr_c = None            # Current most probable class\n",
-        "        \n",
-        "        for y in self.class_values:\n",
-        "            p = np.log(self.probabilities[y])\n",
-        "            for x in self.variables:\n",
-        "                p = p + np.log(self.probabilities[x, row[x].value, y])\n",
-        "            \n",
-        "            if p > curr_p:\n",
-        "                curr_p = p\n",
-        "                curr_c = y\n",
-        "                \n",
-        "        return curr_c, curr_p\n",
-        "        \n",
-        "   \n",
-        "\n",
-        "    def predict(self, data):\n",
-        "        \"\"\"\n",
-        "        Predict class labels for all rows in data.\n",
-        "        \n",
-        "        :param data\n",
-        "            Orange data Table.       \n",
-        "        :return y\n",
-        "            NumPy vector with predicted classes.\n",
-        "        \"\"\"\n",
-        "        \n",
-        "        n = len(data)\n",
-        "        predictions = list()\n",
-        "        confidences = np.zeros((n, ))\n",
-        "        \n",
-        "        for i, row in enumerate(data):\n",
-        "            pred, cf = self.predict_instance(row)\n",
-        "            predictions.append(pred)\n",
-        "            confidences[i] = cf\n",
-        "    \n",
-        "        return predictions, confidences"
+        "print(\"Actual\\tPredicted\")\n",
+        "for i in range(10):\n",
+        "    row = data.iloc[i]\n",
+        "    predictions = dict()\n",
+        "    for y in class_values:\n",
+        "        p = np.log(P[y])\n",
+        "        for x in variables:\n",
+        "            p = p + np.log(P[x, row[x], y])\n",
+        "        predictions[y] = p\n",
+        "    print(max(predictions, key=predictions.get), \"\\t\", row.iloc[-1])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "Praviloma napovednih modelov ne ocenjujemo na podatkih, na katerih smo jih učili. V nadaljevanju bomo uporabili testno množico."
       ]
     },
     {
       "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "As a rule, we do not evaluate predictive models on the data used for training. In the following, we will use the test set."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 15,
       "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Actual\tPredicted\n",
+            "no \t yes\n",
+            "no \t no\n",
+            "no \t no\n",
+            "yes \t yes\n",
+            "yes \t yes\n",
+            "no \t yes\n",
+            "no \t no\n",
+            "yes \t yes\n",
+            "no \t yes\n",
+            "no \t yes\n"
+          ]
+        }
+      ],
+      "source": [
+        "test_data = pd.read_table('../data/titanic-test.tab', skiprows=[1,2])\n",
+        "print(\"Actual\\tPredicted\")\n",
+        "for i in range(10):\n",
+        "    row = test_data.iloc[i]\n",
+        "    predictions = dict()\n",
+        "    for y in class_values:\n",
+        "        p = np.log(P[y])\n",
+        "        for x in variables:\n",
+        "            p = p + np.log(P[x, row[x], y])\n",
+        "        predictions[y] = p\n",
+        "    print(max(predictions, key=predictions.get), \"\\t\", row.iloc[-1])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "Poženimo sedaj naš napovedni model na vseh primerih v testni množici in preštejmo, kolikokrat smo pravilno napovedali."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "Now let’s run our predictive model on all instances in the test set and count how many times we predicted correctly."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 16,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "849"
+            ]
+          },
+          "execution_count": 16,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "results = np.zeros((2, len(test_data)), dtype=int)\n",
+        "map = {'yes': 1, 'no': 0}\n",
+        "for i in range(len(test_data)):\n",
+        "    row = test_data.iloc[i]\n",
+        "    results[0, i] = map[row.iloc[-1]]\n",
+        "    predictions = dict()\n",
+        "    for y in class_values:\n",
+        "        p = np.log(P[y])\n",
+        "        for x in variables:\n",
+        "            p = p + np.log(P[x, row[x], y])\n",
+        "        predictions[y] = p\n",
+        "    results[1, i] = map[max(predictions, key=predictions.get)]\n",
+        "correct = np.equal(results[0], results[1]).sum()\n",
+        "correct"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "Je to dober rezultat ali slab? Število nam samo po sebi nič ne pove, pomembno je v povezavi s celotnim številom primerov. Če bi imeli na primer 850 primerov, bi to bil odličen rezultat. Če bi pa imeli 100.000 primerov, pa porazen. Zato delimo še s številom vseh primerov, da dobimo **napovedno natančnost**, ki jo lahko izrazimo v procentih."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "Is this a good result or a bad one? The number by itself tells us nothingm, what matters is how it relates to the total number of instances. If, for example, we had 850 instances, this would be an excellent result. But if we had 100,000 instances, it would be terrible. That’s why we also divide by the total number of instances to obtain **classification accuracy**, which can be expressed as a percentage."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 17,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "77.1%\n"
+          ]
+        }
+      ],
+      "source": [
+        "print(\"%.1f\" % (correct/len(test_data)*100) + \"%\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "To je že bolj razumljivo: 77,1% primerov smo pravilno napovedali. Običajno modele primerjamo z drugimi modeli. Najenostavnejši klasifikacijski model je večinski klasifikator. Kot nakazuje že samo ime, vsem primerom napove vrednost razreda, ki se največkrat pojavi. V našem primeru so to potniki, ki niso preživeli."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "This is already easier to understand: we correctly predicted 77.1% of the cases. Usually we compare models with other models. The simplest classification model is the majority classifier. As the name suggests, it predicts for all instances the value of the class that appears most frequently. In our case, these are the passengers who did not survive."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 18,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "survived\n",
+              "no     748\n",
+              "yes    353\n",
+              "Name: count, dtype: int64"
+            ]
+          },
+          "execution_count": 18,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "test_data['survived'].value_counts() "
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 19,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "67.9%\n"
+          ]
+        }
+      ],
+      "source": [
+        "majority = np.zeros(len(test_data), dtype=int)\n",
+        "correct = np.equal(results[0], majority).sum()\n",
+        "print(\"%.1f\" % (correct/len(test_data)*100) + \"%\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "Vidimo, da ima množinski klasifikator samo manj kot 10% slabšo napovedno natačnost. A to je pravzaprav samo eden od načinov ocenjevanja uspešnosti klasifikatorjev.\n",
+        "\n",
+        "Vsak napovedani primer primerjamo s pripadajočim resničnim razredom. Štirje možni izidi primerjave so naslednji: \n",
+        "\n",
+        "<table cellspacing=\"0\" cellpadding=\"8\" style=\"border-collapse: collapse; text-align: center;\">\n",
+        "    <tr>\n",
+        "        <td  colspan=\"2\" rowspan=\"2\" style=\"border:0px\"></td>\n",
+        "        <th colspan=\"2\">Dejanski razred</th>\n",
+        "    </tr>\n",
+        "    <tr>\n",
+        "        <th>Negativni</th>\n",
+        "        <th>Pozitivni</th>\n",
+        "    </tr>\n",
+        "    <tr>\n",
+        "        <th rowspan=\"2\" style=\"\n",
+        "            writing-mode: vertical-rl;\n",
+        "            transform: rotate(180deg);\n",
+        "            padding: 10px;\n",
+        "        \">\n",
+        "            Napovedi\n",
+        "        </th>\n",
+        "        <th style=\"\n",
+        "            writing-mode: vertical-rl;\n",
+        "            transform: rotate(180deg);\">Negativni</th>\n",
+        "        <td>TN: True negatives<br>(pravilno napovedani negativni primeri)</td>\n",
+        "        <td>FN: False negatives<br>(napačno napovedani pozitivni primeri)</td>\n",
+        "    </tr>\n",
+        "    <tr>\n",
+        "        <th style=\"\n",
+        "            writing-mode: vertical-rl;\n",
+        "            transform: rotate(180deg);\">Pozitivni</th>\n",
+        "        <td>FP: False positives<br>(napačno napovedani negativni primeri)</td>\n",
+        "        <td>TP: True positives<br>(pravilno napovedani pozitivni primeri)</td>\n",
+        "    </tr>\n",
+        "</table>"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "We can see that the majority classifier has only a bit less than 10% worse predictive accuracy. But this is actually only one of the ways to evaluate the performance of classifiers.\n",
+        "\n",
+        "For each predicted instance, we compare it with its corresponding true class. The four possible outcomes of this comparison are the following:\n",
+        "\n",
+        "<table cellspacing=\"0\" cellpadding=\"8\" style=\"border-collapse: collapse; text-align: center;\">\n",
+        "    <tr>\n",
+        "        <td  colspan=\"2\" rowspan=\"2\" style=\"border:0px\"></td>\n",
+        "        <th colspan=\"2\">Actual class</th>\n",
+        "    </tr>\n",
+        "    <tr>\n",
+        "        <th>Negative</th>\n",
+        "        <th>Positive</th>\n",
+        "    </tr>\n",
+        "    <tr>\n",
+        "        <th rowspan=\"2\" style=\"\n",
+        "            writing-mode: vertical-rl;\n",
+        "            transform: rotate(180deg);\n",
+        "            padding: 10px;\n",
+        "        \">\n",
+        "            Prediction\n",
+        "        </th>\n",
+        "        <th style=\"\n",
+        "            writing-mode: vertical-rl;\n",
+        "            transform: rotate(180deg);\">Negative</th>\n",
+        "        <td>TN: True negatives<br>(correctly predicted negative examples)</td>\n",
+        "        <td>FN: False negatives<br>(wrongly predicted positive examples)</td>\n",
+        "    </tr>\n",
+        "    <tr>\n",
+        "        <th style=\"\n",
+        "            writing-mode: vertical-rl;\n",
+        "            transform: rotate(180deg);\">Positive</th>\n",
+        "        <td>FP: False positives<br>(wrongly predicted negative examples)</td>\n",
+        "        <td>TP: True positives<br>(correctly predicted positive examples)</td>\n",
+        "    </tr>\n",
+        "</table>"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 20,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "array([[688., 192.],\n",
+              "       [ 60., 161.]])"
+            ]
+          },
+          "execution_count": 20,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "table = np.zeros((2,2))\n",
+        "for i in range(len(test_data)):\n",
+        "    table[results[1,i], results[0,i]] += 1\n",
+        "[[TN, FN],[FP,TP]] = table\n",
+        "table"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 21,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "array([[748., 353.],\n",
+              "       [  0.,   0.]])"
+            ]
+          },
+          "execution_count": 21,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "table = np.zeros((2,2))\n",
+        "for i in range(len(test_data)):\n",
+        "    table[majority[i], results[0,i]] += 1\n",
+        "table"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "### Delež pravilno razvrščenih razredov (ang. classification accuracy)\n",
+        "\n",
+        "$$ca = \\frac{TP + TN}{TP + TN + FP + FN}$$\n",
+        "\n",
+        "<font color=\"green\">Prednosti</font>:\n",
+        "* Enostaven izračun, jasna interpretacija\n",
+        "* Uporabna mera za poljubno število razredov\n",
+        "\n",
+        "<font color=\"red\">Slabosti</font>:\n",
+        "* Lahko zavaja pri neuravnoteženih porazdelitvah razredov\n",
+        "\n",
+        "<br/>"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "### Ratio of correctly classified classes (classification accuracy)\n",
+        "\n",
+        "$$ca = \\frac{TP + TN}{TP + TN + FP + FN}$$\n",
+        "\n",
+        "<font color=\"green\">Pros</font>:\n",
+        "* Simple calculation, clear interpretation\n",
+        "* Useful measure for any number of classes\n",
+        "\n",
+        "<font color=\"red\">Cons</font>:\n",
+        "* It can be misleading with unbalanced class distributions"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 22,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "77.1%\n"
+          ]
+        }
+      ],
+      "source": [
+        "ca = (TP+TN) / (TP+TN+FP+FN)\n",
+        "print(\"%.1f\" % (ca*100) + \"%\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "### Natančnost, priklic (ang. precision, recall)\n",
+        "\n",
+        "Natančnost nam pove delež dejansko pozitivnih primerov med pozitivnimi napovedmi.\n",
+        "\n",
+        "$$ p = \\frac{TP}{TP + FP} $$\n",
+        "\n",
+        "Priklic nam pove delež dejansko pozitivnih primerov, ki jih je naša metoda odkrila.\n",
+        "\n",
+        "$$ r = \\frac{TP}{TP + FN} $$\n",
+        "\n",
+        "<font color=\"green\">Prednosti</font>:\n",
+        "* Enostaven izračun, jasna interpretacija\n",
+        "* Ločitev obeh tipov napak (napačno pozitivni in napačno negativni primeri)\n",
+        "* Uporabna tudi pri neuravnoteženih porazdelitvah razredov\n",
+        "\n",
+        "<font color=\"red\">Slabosti</font>:\n",
+        "* Uporabno pretežno za klasifikacijo v dva razreda\n",
+        "* Težko povzeti obe meri ; približek je F1-vrednost (ang. F1-score)\n",
+        "$$ F1 = 2 \\frac{p \\cdot r}{p + r} $$"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "### Precision, recall\n",
+        "\n",
+        "Precision tells us the proportion of truly positive cases among all positive predictions.\n",
+        "\n",
+        "$$ p = \\frac{TP}{TP + FP} $$\n",
+        "\n",
+        "Recall tells us the proportion of truly positive cases that our method successfully identified.\n",
+        "\n",
+        "$$ r = \\frac{TP}{TP + FN} $$\n",
+        "\n",
+        "<font color=\"green\">Pros</font>:\n",
+        "* Simple calculation, clear interpretation\n",
+        "* Separation of both types of errors (incorrectly positive and wrongly negative examples)\n",
+        "* Also applicable for unbalanced classroom distributions\n",
+        "\n",
+        "<font color=\"red\">Cons</font>:\n",
+        "* Applicable predominantly for classification in two classes\n",
+        "* It is difficult to summarize both measures; the approximation is F1-value (F1-score)\n",
+        "$$ F1 = 2 \\frac{p \\cdot r}{p + r} $$"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 23,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "prec.:\t72.9%\n",
+            "recall:\t45.6%\n",
+            "F1:\t56.1%\n"
+          ]
+        }
+      ],
+      "source": [
+        "p = TP / (TP+FP)\n",
+        "print(\"prec.:\\t%.1f\" % (p*100) + \"%\")\n",
+        "r = TP / (TP+FN)\n",
+        "print(\"recall:\\t%.1f\" % (r*100) + \"%\")\n",
+        "F1 = 2*(p*r)/(p+r)\n",
+        "print(\"F1:\\t%.1f\" % (F1*100) + \"%\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "<p style=\"text-align: center\">\n",
+        "<img width=\"400\" src=\"../slike/Precisionrecall.png\"></img>\n",
+        "</p>"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "Kakšna pa je uspešnost večinskega klasifikatorja? Ker je $TP=0$, je tudi priklic $0%$, in ker je tudi $FP=0$ natančnosti ne moremo izračunati.\n",
+        "\n",
+        "#### Vprašanje 7-1-1\n",
+        "\n",
+        "Zamenjaj pozitivni in negativni razred in ponovno oceni uspešnost obeh modelov."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
       "source": [
-        "Re\u0161itev je dostopna na: 205-2.ipynb"
+        "And what is the performance of the majority classifier? Since $TP = 0$, the recall is also $0%$, and because $FP = 0$ as well, we cannot compute the precision.\n",
+        "\n",
+        "#### Question 7-1-1\n",
+        "\n",
+        "Swap the positive and negative class and evaluate the performance of both models again.\n"
       ]
     },
     {
       "cell_type": "code",
-      "execution_count": 11,
+      "execution_count": null,
       "metadata": {},
       "outputs": [],
-      "source": [
-        "%run 205-2.ipynb"
-      ]
+      "source": []
     },
     {
       "cell_type": "markdown",
@@ -878,7 +1548,7 @@
         ]
       },
       "source": [
-        "##  Uporaba klasifikatorja"
+        "[Odgovor](207-1.ipynb#Odgovor-7-1-1)"
       ]
     },
     {
@@ -889,7 +1559,7 @@
         ]
       },
       "source": [
-        "## Using a classifier"
+        "[Answer](207-1.ipynb#Answer-7-1-1)"
       ]
     },
     {
@@ -900,7 +1570,13 @@
         ]
       },
       "source": [
-        "Primer uporabe na podatkih potnikov ladje <i><a href=\"https://www.kaggle.com/c/titanic\">Titanic</a></i>."
+        "## Odločitvena drevesa\n",
+        "\n",
+        "Po obravnavi Naivnega Bayesovega klasifikatorja (modela, ki hkrati upošteva vse atribute in njihove prispevke združuje verjetnostno, ob predpostavki pogojne neodvisnosti) si zdaj ogledamo metodo, ki se klasifikacije loteva povsem drugače: odločitvena drevesa. Namesto izračuna skupnih verjetnosti odločitveno drevo gradi zaporedje preprostih, s pravili opredeljenih odločitev.\n",
+        "\n",
+        "Vsaka razvejitvena točka v drevesu temelji na enem samem atributu in odgovarja na vprašanje, kot je »Ali je višina večja od 170 cm?« ali »Ali je vrednost enaka ‘da’?«. Z zastavljanjem enega vprašanja naenkrat in postopnim razdeljevanjem podatkov drevo ustvarja veje, ki primerke ločujejo v vedno bolj homogene skupine. Na koncu vsak primer pristane v listu drevesa, ki predstavlja napovedani razred.\n",
+        "\n",
+        "Ena od ključnih prednosti odločitvenih dreves je njihova razložljivost. Za razliko od Naivnega Bayesa, ki informacije vseh atributov združi naenkrat na način, ki ga včasih ni enostavno slediti, drevo celoten sklepni postopek prikaže povsem jasno. Vsaka odločitev, vsaka razvejitev in vsak sklep sta vidna, zato je metoda posebej uporabna, kadar sta preglednost in interpretacija enako pomembni kot natančnost."
       ]
     },
     {
@@ -911,89 +1587,508 @@
         ]
       },
       "source": [
-        "An example of use on passenger data from <i><a href=\"https://www.kaggle.com/c/titanic\">Titanic</a></i>."
+        "## Decision Trees\n",
+        "\n",
+        "After exploring Naive Bayes, a model that evaluates all attributes simultaneously and combines their contributions probabilistically under the assumption of conditional independence, we now turn to a method that approaches classification quite differently: decision trees. Instead of computing joint probabilities, a decision tree builds a sequence of simple, rule-based decisions.\n",
+        "\n",
+        "Each split in the tree is based on a single attribute and answers a question such as “Is the height greater than 170 cm?” or “Is the value equal to ‘yes’?”. By asking one question at a time and dividing the dataset step by step, the tree creates branches that separate examples into increasingly homogeneous groups. Eventually, each sample reaches a leaf node, which corresponds to a predicted class.\n",
+        "\n",
+        "One of the key advantages of decision trees is their interpretability. Unlike Naive Bayes, which combines information from all attributes at once in a way that’s sometimes difficult to trace, a tree shows its entire reasoning path explicitly. Every decision, every split, and every conclusion can be visualized and understood, making the method especially useful when transparency matters as much as accuracy."
       ]
     },
     {
       "cell_type": "code",
-      "execution_count": 12,
+      "execution_count": 24,
       "metadata": {},
       "outputs": [
         {
           "data": {
+            "text/html": [
+              "<style>#sk-container-id-1 {\n",
+              "  /* Definition of color scheme common for light and dark mode */\n",
+              "  --sklearn-color-text: black;\n",
+              "  --sklearn-color-line: gray;\n",
+              "  /* Definition of color scheme for unfitted estimators */\n",
+              "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
+              "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
+              "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
+              "  --sklearn-color-unfitted-level-3: chocolate;\n",
+              "  /* Definition of color scheme for fitted estimators */\n",
+              "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
+              "  --sklearn-color-fitted-level-1: #d4ebff;\n",
+              "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
+              "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
+              "\n",
+              "  /* Specific color for light theme */\n",
+              "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
+              "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
+              "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
+              "  --sklearn-color-icon: #696969;\n",
+              "\n",
+              "  @media (prefers-color-scheme: dark) {\n",
+              "    /* Redefinition of color scheme for dark theme */\n",
+              "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
+              "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
+              "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
+              "    --sklearn-color-icon: #878787;\n",
+              "  }\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 {\n",
+              "  color: var(--sklearn-color-text);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 pre {\n",
+              "  padding: 0;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 input.sk-hidden--visually {\n",
+              "  border: 0;\n",
+              "  clip: rect(1px 1px 1px 1px);\n",
+              "  clip: rect(1px, 1px, 1px, 1px);\n",
+              "  height: 1px;\n",
+              "  margin: -1px;\n",
+              "  overflow: hidden;\n",
+              "  padding: 0;\n",
+              "  position: absolute;\n",
+              "  width: 1px;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-dashed-wrapped {\n",
+              "  border: 1px dashed var(--sklearn-color-line);\n",
+              "  margin: 0 0.4em 0.5em 0.4em;\n",
+              "  box-sizing: border-box;\n",
+              "  padding-bottom: 0.4em;\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-container {\n",
+              "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
+              "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
+              "     so we also need the `!important` here to be able to override the\n",
+              "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
+              "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
+              "  display: inline-block !important;\n",
+              "  position: relative;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-text-repr-fallback {\n",
+              "  display: none;\n",
+              "}\n",
+              "\n",
+              "div.sk-parallel-item,\n",
+              "div.sk-serial,\n",
+              "div.sk-item {\n",
+              "  /* draw centered vertical line to link estimators */\n",
+              "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
+              "  background-size: 2px 100%;\n",
+              "  background-repeat: no-repeat;\n",
+              "  background-position: center center;\n",
+              "}\n",
+              "\n",
+              "/* Parallel-specific style estimator block */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel-item::after {\n",
+              "  content: \"\";\n",
+              "  width: 100%;\n",
+              "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
+              "  flex-grow: 1;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel {\n",
+              "  display: flex;\n",
+              "  align-items: stretch;\n",
+              "  justify-content: center;\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "  position: relative;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel-item {\n",
+              "  display: flex;\n",
+              "  flex-direction: column;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel-item:first-child::after {\n",
+              "  align-self: flex-end;\n",
+              "  width: 50%;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel-item:last-child::after {\n",
+              "  align-self: flex-start;\n",
+              "  width: 50%;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel-item:only-child::after {\n",
+              "  width: 0;\n",
+              "}\n",
+              "\n",
+              "/* Serial-specific style estimator block */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-serial {\n",
+              "  display: flex;\n",
+              "  flex-direction: column;\n",
+              "  align-items: center;\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "  padding-right: 1em;\n",
+              "  padding-left: 1em;\n",
+              "}\n",
+              "\n",
+              "\n",
+              "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
+              "clickable and can be expanded/collapsed.\n",
+              "- Pipeline and ColumnTransformer use this feature and define the default style\n",
+              "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
+              "*/\n",
+              "\n",
+              "/* Pipeline and ColumnTransformer style (default) */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-toggleable {\n",
+              "  /* Default theme specific background. It is overwritten whether we have a\n",
+              "  specific estimator or a Pipeline/ColumnTransformer */\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "}\n",
+              "\n",
+              "/* Toggleable label */\n",
+              "#sk-container-id-1 label.sk-toggleable__label {\n",
+              "  cursor: pointer;\n",
+              "  display: block;\n",
+              "  width: 100%;\n",
+              "  margin-bottom: 0;\n",
+              "  padding: 0.5em;\n",
+              "  box-sizing: border-box;\n",
+              "  text-align: center;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 label.sk-toggleable__label-arrow:before {\n",
+              "  /* Arrow on the left of the label */\n",
+              "  content: \"▸\";\n",
+              "  float: left;\n",
+              "  margin-right: 0.25em;\n",
+              "  color: var(--sklearn-color-icon);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {\n",
+              "  color: var(--sklearn-color-text);\n",
+              "}\n",
+              "\n",
+              "/* Toggleable content - dropdown */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-toggleable__content {\n",
+              "  max-height: 0;\n",
+              "  max-width: 0;\n",
+              "  overflow: hidden;\n",
+              "  text-align: left;\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-toggleable__content.fitted {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-0);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-toggleable__content pre {\n",
+              "  margin: 0.2em;\n",
+              "  border-radius: 0.25em;\n",
+              "  color: var(--sklearn-color-text);\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-toggleable__content.fitted pre {\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-0);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
+              "  /* Expand drop-down */\n",
+              "  max-height: 200px;\n",
+              "  max-width: 100%;\n",
+              "  overflow: auto;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
+              "  content: \"▾\";\n",
+              "}\n",
+              "\n",
+              "/* Pipeline/ColumnTransformer-specific style */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
+              "  color: var(--sklearn-color-text);\n",
+              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
+              "  background-color: var(--sklearn-color-fitted-level-2);\n",
+              "}\n",
+              "\n",
+              "/* Estimator-specific style */\n",
+              "\n",
+              "/* Colorize estimator box */\n",
+              "#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-2);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-label label.sk-toggleable__label,\n",
+              "#sk-container-id-1 div.sk-label label {\n",
+              "  /* The background is the default theme color */\n",
+              "  color: var(--sklearn-color-text-on-default-background);\n",
+              "}\n",
+              "\n",
+              "/* On hover, darken the color of the background */\n",
+              "#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {\n",
+              "  color: var(--sklearn-color-text);\n",
+              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
+              "}\n",
+              "\n",
+              "/* Label box, darken color on hover, fitted */\n",
+              "#sk-container-id-1 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
+              "  color: var(--sklearn-color-text);\n",
+              "  background-color: var(--sklearn-color-fitted-level-2);\n",
+              "}\n",
+              "\n",
+              "/* Estimator label */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-label label {\n",
+              "  font-family: monospace;\n",
+              "  font-weight: bold;\n",
+              "  display: inline-block;\n",
+              "  line-height: 1.2em;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-label-container {\n",
+              "  text-align: center;\n",
+              "}\n",
+              "\n",
+              "/* Estimator-specific */\n",
+              "#sk-container-id-1 div.sk-estimator {\n",
+              "  font-family: monospace;\n",
+              "  border: 1px dotted var(--sklearn-color-border-box);\n",
+              "  border-radius: 0.25em;\n",
+              "  box-sizing: border-box;\n",
+              "  margin-bottom: 0.5em;\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-estimator.fitted {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-0);\n",
+              "}\n",
+              "\n",
+              "/* on hover */\n",
+              "#sk-container-id-1 div.sk-estimator:hover {\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-estimator.fitted:hover {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-2);\n",
+              "}\n",
+              "\n",
+              "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
+              "\n",
+              "/* Common style for \"i\" and \"?\" */\n",
+              "\n",
+              ".sk-estimator-doc-link,\n",
+              "a:link.sk-estimator-doc-link,\n",
+              "a:visited.sk-estimator-doc-link {\n",
+              "  float: right;\n",
+              "  font-size: smaller;\n",
+              "  line-height: 1em;\n",
+              "  font-family: monospace;\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "  border-radius: 1em;\n",
+              "  height: 1em;\n",
+              "  width: 1em;\n",
+              "  text-decoration: none !important;\n",
+              "  margin-left: 1ex;\n",
+              "  /* unfitted */\n",
+              "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
+              "  color: var(--sklearn-color-unfitted-level-1);\n",
+              "}\n",
+              "\n",
+              ".sk-estimator-doc-link.fitted,\n",
+              "a:link.sk-estimator-doc-link.fitted,\n",
+              "a:visited.sk-estimator-doc-link.fitted {\n",
+              "  /* fitted */\n",
+              "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
+              "  color: var(--sklearn-color-fitted-level-1);\n",
+              "}\n",
+              "\n",
+              "/* On hover */\n",
+              "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
+              ".sk-estimator-doc-link:hover,\n",
+              "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
+              ".sk-estimator-doc-link:hover {\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-3);\n",
+              "  color: var(--sklearn-color-background);\n",
+              "  text-decoration: none;\n",
+              "}\n",
+              "\n",
+              "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
+              ".sk-estimator-doc-link.fitted:hover,\n",
+              "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
+              ".sk-estimator-doc-link.fitted:hover {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-3);\n",
+              "  color: var(--sklearn-color-background);\n",
+              "  text-decoration: none;\n",
+              "}\n",
+              "\n",
+              "/* Span, style for the box shown on hovering the info icon */\n",
+              ".sk-estimator-doc-link span {\n",
+              "  display: none;\n",
+              "  z-index: 9999;\n",
+              "  position: relative;\n",
+              "  font-weight: normal;\n",
+              "  right: .2ex;\n",
+              "  padding: .5ex;\n",
+              "  margin: .5ex;\n",
+              "  width: min-content;\n",
+              "  min-width: 20ex;\n",
+              "  max-width: 50ex;\n",
+              "  color: var(--sklearn-color-text);\n",
+              "  box-shadow: 2pt 2pt 4pt #999;\n",
+              "  /* unfitted */\n",
+              "  background: var(--sklearn-color-unfitted-level-0);\n",
+              "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
+              "}\n",
+              "\n",
+              ".sk-estimator-doc-link.fitted span {\n",
+              "  /* fitted */\n",
+              "  background: var(--sklearn-color-fitted-level-0);\n",
+              "  border: var(--sklearn-color-fitted-level-3);\n",
+              "}\n",
+              "\n",
+              ".sk-estimator-doc-link:hover span {\n",
+              "  display: block;\n",
+              "}\n",
+              "\n",
+              "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
+              "\n",
+              "#sk-container-id-1 a.estimator_doc_link {\n",
+              "  float: right;\n",
+              "  font-size: 1rem;\n",
+              "  line-height: 1em;\n",
+              "  font-family: monospace;\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "  border-radius: 1rem;\n",
+              "  height: 1rem;\n",
+              "  width: 1rem;\n",
+              "  text-decoration: none;\n",
+              "  /* unfitted */\n",
+              "  color: var(--sklearn-color-unfitted-level-1);\n",
+              "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 a.estimator_doc_link.fitted {\n",
+              "  /* fitted */\n",
+              "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
+              "  color: var(--sklearn-color-fitted-level-1);\n",
+              "}\n",
+              "\n",
+              "/* On hover */\n",
+              "#sk-container-id-1 a.estimator_doc_link:hover {\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-3);\n",
+              "  color: var(--sklearn-color-background);\n",
+              "  text-decoration: none;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 a.estimator_doc_link.fitted:hover {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-3);\n",
+              "}\n",
+              "</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>DecisionTreeClassifier(max_depth=3, random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;DecisionTreeClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.4/modules/generated/sklearn.tree.DecisionTreeClassifier.html\">?<span>Documentation for DecisionTreeClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>DecisionTreeClassifier(max_depth=3, random_state=42)</pre></div> </div></div></div></div>"
+            ],
             "text/plain": [
-              "{'no': 0.6745454545454546,\n",
-              " 'yes': 0.32545454545454544,\n",
-              " ('status', 'third', 'no'): 0.3450134770889488,\n",
-              " ('status', 'second', 'no'): 0.12398921832884097,\n",
-              " ('status', 'crew', 'no'): 0.4568733153638814,\n",
-              " ('status', 'first', 'no'): 0.07412398921832884,\n",
-              " ('age', 'adult', 'no'): 0.9663072776280324,\n",
-              " ('age', 'child', 'no'): 0.03369272237196765,\n",
-              " ('sex', 'male', 'no'): 0.9110512129380054,\n",
-              " ('sex', 'female', 'no'): 0.0889487870619946,\n",
-              " ('status', 'third', 'yes'): 0.24581005586592178,\n",
-              " ('status', 'second', 'yes'): 0.17039106145251395,\n",
-              " ('status', 'crew', 'yes'): 0.29329608938547486,\n",
-              " ('status', 'first', 'yes'): 0.2905027932960894,\n",
-              " ('age', 'adult', 'yes'): 0.9162011173184358,\n",
-              " ('age', 'child', 'yes'): 0.08379888268156424,\n",
-              " ('sex', 'male', 'yes'): 0.48044692737430167,\n",
-              " ('sex', 'female', 'yes'): 0.5195530726256983}"
+              "DecisionTreeClassifier(max_depth=3, random_state=42)"
             ]
           },
-          "execution_count": 12,
+          "execution_count": 24,
           "metadata": {},
           "output_type": "execute_result"
         }
       ],
       "source": [
-        "model = NaiveBayes()\n",
-        "model.fit(data)\n",
-        "model.probabilities"
+        "from sklearn.tree import DecisionTreeClassifier, plot_tree\n",
+        "import matplotlib.pyplot as plt\n",
+        "\n",
+        "data = pd.read_table('../data/sportniki.tab')\n",
+        "\n",
+        "X = data[[\"visina\"]].apply(lambda col: col.astype(\"category\").cat.codes)\n",
+        "sport_cat = data[\"sport\"].astype(\"category\")\n",
+        "y = sport_cat.cat.codes\n",
+        "\n",
+        "tree = DecisionTreeClassifier(max_depth=3, random_state=42)\n",
+        "tree.fit(X, y)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "Drevo lahko tudi izrišemo. Ker smo kategorije višine v predhodnem koraku spremenili v številčne vrednosti, izpišimo še pretvorbo za lažje razumevanje modela."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "We can also plot the tree. Since we converted the height categories into numerical values in the previous step, let’s also print the mapping to make the model easier to understand."
       ]
     },
     {
       "cell_type": "code",
-      "execution_count": 13,
+      "execution_count": 25,
       "metadata": {},
       "outputs": [
         {
           "name": "stdout",
           "output_type": "stream",
           "text": [
-            "Status=third, age=adult sex=male\n",
-            "Actual class=no, predicted class=no confidence=-1.58532\n",
-            "Status=second, age=adult sex=female\n",
-            "Actual class=no, predicted class=yes confidence=-3.63450\n",
-            "Status=crew, age=adult sex=male\n",
-            "Actual class=no, predicted class=no confidence=-1.30449\n",
-            "Status=crew, age=adult sex=male\n",
-            "Actual class=no, predicted class=no confidence=-1.30449\n",
-            "Status=third, age=adult sex=male\n",
-            "Actual class=no, predicted class=no confidence=-1.58532\n",
-            "Status=second, age=adult sex=male\n",
-            "Actual class=no, predicted class=no confidence=-2.60871\n",
-            "Status=crew, age=adult sex=male\n",
-            "Actual class=no, predicted class=no confidence=-1.30449\n",
-            "Status=second, age=adult sex=male\n",
-            "Actual class=no, predicted class=no confidence=-2.60871\n",
-            "Status=third, age=adult sex=male\n",
-            "Actual class=yes, predicted class=no confidence=-1.58532\n",
-            "Status=third, age=adult sex=male\n",
-            "Actual class=no, predicted class=no confidence=-1.58532\n"
+            "{0: 'nizek', 1: 'srednji', 2: 'visok'}\n"
           ]
+        },
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 1000x600 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
         }
       ],
       "source": [
-        "predictions, confidences = model.predict(data)\n",
-        "\n",
-        "for i in range(10):\n",
-        "    row = data.iloc[i]\n",
-        "    p = predictions[i]\n",
-        "    c = confidences[i]\n",
-        "    print(\"Status=%s, age=%s sex=%s\" % (row['status'], row['age'], row['sex']))\n",
-        "    print(\"Actual class=%s, predicted class=%s confidence=%.5f\" % (row['survived'], p, c))"
+        "print(dict(enumerate(data[\"visina\"].astype(\"category\").cat.categories)))\n",
+        "plt.figure(figsize=(10, 6))\n",
+        "plot_tree(\n",
+        "    tree,\n",
+        "    feature_names=[\"visina\"],\n",
+        "    class_names=data[\"sport\"].astype(\"category\").cat.categories,\n",
+        "    filled=True,\n",
+        "    rounded=True\n",
+        ")\n",
+        "plt.show()"
       ]
     },
     {
@@ -1004,9 +2099,7 @@
         ]
       },
       "source": [
-        "## Ocenjevanje uspe\u0161nosti klasifikacije\n",
-        "\n",
-        "Za ocenjevanje uspe\u0161nosti klasifikacije vsak napovedani primer primerjamo s pripadajo\u010dim resni\u010dnim razredom. \u0160tirje mo\u017eni izidi primerjave so naslednji: "
+        "Model lahko uporabimo za napovedovanje na novih primerih. Mark je še vedno srednje rasti (vrednost $1$)."
       ]
     },
     {
@@ -1017,9 +2110,28 @@
         ]
       },
       "source": [
-        "## Assessing the performance of the classification\n",
-        "\n",
-        "In order to evaluate the success of the classification, we compare each predicted example with the corresponding real class. The four possible outcomes of the comparison are as follows:"
+        "We can use the model to make predictions on new cases. Mark is still of medium height (value $1$)."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "array([2], dtype=int8)"
+            ]
+          },
+          "execution_count": 26,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "mark = pd.DataFrame({\"visina\": [1]})\n",
+        "tree.predict(mark)"
       ]
     },
     {
@@ -1030,23 +2142,7 @@
         ]
       },
       "source": [
-        "<table>\n",
-        "<tr>\n",
-        "<td>\n",
-        "<ul>\n",
-        "<li>TP: True positives (pravilno napovedani pozitivni primeri)</li>\n",
-        "<li>FP: False positives (napa\u010dno napovedani negativni primeri)</li>\n",
-        "<li>TN: True negatives (pravilno napovedani negativni primeri)</li>\n",
-        "<li>FN: False negatives (napa\u010dno napovedani pozitini primeri)</li>\n",
-        "</ul> \n",
-        "\n",
-        "<br/>\n",
-        "<img src=\"slike/type12_error.jpeg\" width=400/>\n",
-        "\n",
-        "</td>\n",
-        "<td><img width=\"400\" src=\"slike/Precisionrecall.png\"></img><td>\n",
-        "<tr/>\n",
-        "<table>"
+        "Sama številka nam ne pove veliko, zato jo pretvorimo v ime športa."
       ]
     },
     {
@@ -1057,23 +2153,31 @@
         ]
       },
       "source": [
-        "<table>\n",
-        "<tr>\n",
-        "<td>\n",
-        "<ul>\n",
-        "<li>TP: True Positives (correctly predicted positive examples) </li>\n",
-        "<li>FP: False positives (wrongly predicted negative examples) </li>\n",
-        "<li>TN: True Negatives (correctly predicted negative examples) </li>\n",
-        "<li>FN: False negatives (wrongly predicted positive examples) </li>\n",
-        "</ul>\n",
-        "\n",
-        "<br/>\n",
-        "<img src=\"slike/type12_error.jpeg\" width=400/>\n",
+        "The number alone doesn’t tell us much, so we convert it back into the sport’s name."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "Index(['nogomet'], dtype='object')"
+            ]
+          },
+          "execution_count": 27,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "class_names = sport_cat.cat.categories\n",
+        "pred = tree.predict(mark) \n",
+        "pred_label = class_names[pred]\n",
         "\n",
-        "</td>\n",
-        "<td><img width=\"400\" src=\"slike/Precisionrecall.png\"></img><td>\n",
-        "<tr/>\n",
-        "<table>"
+        "pred_label"
       ]
     },
     {
@@ -1084,18 +2188,7 @@
         ]
       },
       "source": [
-        "### Dele\u017e pravilno razvr\u0161\u010denih razredov (ang. classification accuracy)\n",
-        "\n",
-        "$$ca = \\frac{TP + TN}{TP + TN + FP + FN}$$\n",
-        "\n",
-        "<font color=\"green\">Prednosti</font>:\n",
-        "* Enostaven izra\u010dun, jasna interpretacija\n",
-        "* Uporabna mera za poljubno \u0161tevilo razredov\n",
-        "\n",
-        "<font color=\"red\">Slabosti</font>:\n",
-        "* Lahko zavaja pri neuravnote\u017eenih porazdelitvah razredov\n",
-        "\n",
-        "<br/>"
+        "Sedaj bomo na podatkih o Titaniku naučili odločitveno drevo. Uporabimo atribute `status`, `age` in `sex` za napoved `survived`. Po učenju si oglejmo izrisano drevo."
       ]
     },
     {
@@ -1106,16 +2199,49 @@
         ]
       },
       "source": [
-        "### Ratio of correctly classified classes (classification accuracy)\n",
-        "\n",
-        "$$ca = \\frac{TP + TN}{TP + TN + FP + FN}$$\n",
-        "\n",
-        "<font color=\"green\">Pros</font>:\n",
-        "* Simple calculation, clear interpretation\n",
-        "* Useful measure for any number of classes\n",
-        "\n",
-        "<font color=\"red\">Cons</font>:\n",
-        "* It can be misleading with unbalanced class distributions"
+        "Now let’s train a decision tree on the Titanic dataset. Use the features `status`, `age`, and `sex` to predict `survived`. After training, visualize the resulting tree."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 28,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 1600x1000 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "data = pd.read_table('../data/titanic-training.tab', skiprows=[1,2])\n",
+        "data[\"status\"] = data[\"status\"].astype(\"category\").cat.codes\n",
+        "data[\"age\"] = data[\"age\"].astype(\"category\").cat.codes   # child=1, adult=0\n",
+        "data[\"sex\"] = data[\"sex\"].astype(\"category\").cat.codes   # male=1, female=0\n",
+        "\n",
+        "X = data[[\"status\", \"age\", \"sex\"]]\n",
+        "y = data[\"survived\"]\n",
+        "\n",
+        "tree = DecisionTreeClassifier(\n",
+        "    criterion=\"entropy\",\n",
+        "    max_depth=4,\n",
+        "    random_state=42\n",
+        ")\n",
+        "tree.fit(X, y)\n",
+        "\n",
+        "plt.figure(figsize=(16, 10))\n",
+        "plot_tree(\n",
+        "    tree,\n",
+        "    feature_names=X.columns,\n",
+        "    class_names=[\"died\", \"survived\"],\n",
+        "    filled=True,\n",
+        ")\n",
+        "plt.show()\n"
       ]
     },
     {
@@ -1126,21 +2252,9 @@
         ]
       },
       "source": [
-        "### Natan\u010dnost, priklic (ang. precision, recall)\n",
-        "\n",
-        "$$ p = \\frac{TP}{TP + FP} $$\n",
-        "\n",
-        "$$ r = \\frac{TP}{TP + FN} $$\n",
-        "\n",
-        "<font color=\"green\">Prednosti</font>:\n",
-        "* Enostaven izra\u010dun, jasna interpretacija\n",
-        "* Lo\u010ditev obeh tipov napak (napa\u010dno pozitivni in napa\u010dno negativni primeri)\n",
-        "* Uporabna tudi pri neuravnote\u017eenih porazdelitvah razredov\n",
+        "### Vprašanje 7-1-2\n",
         "\n",
-        "<font color=\"red\">Slabosti</font>:\n",
-        "* Uporabno prete\u017eno za klasifikacijo v dva razreda\n",
-        "* Te\u017eko povzeti obe meri ; pribli\u017eek je F1-vrednost (ang. F1-score)\n",
-        "$$ F1 = 2 \\frac{p \\cdot r}{p + r} $$"
+        "Na testni množici ocenite uspešnost klasifikatorja."
       ]
     },
     {
@@ -1151,23 +2265,18 @@
         ]
       },
       "source": [
-        "### Precision, recall\n",
-        "\n",
-        "$$ p = \\frac{TP}{TP + FP} $$\n",
-        "\n",
-        "$$ r = \\frac{TP}{TP + FN} $$\n",
-        "\n",
-        "<font color=\"green\">Pros</font>:\n",
-        "* Simple calculation, clear interpretation\n",
-        "* Separation of both types of errors (incorrectly positive and wrongly negative examples)\n",
-        "* Also applicable for unbalanced classroom distributions\n",
+        "### Question 7-1-2\n",
         "\n",
-        "<font color=\"red\">Cons</font>:\n",
-        "* Applicable predominantly for classification in two classes\n",
-        "* It is difficult to summarize both measures; the approximation is F1-value (F1-score)\n",
-        "$$ F1 = 2 \\frac{p \\cdot r}{p + r} $$"
+        "Evaluate the performance of the classifier on the test set."
       ]
     },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": []
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -1176,7 +2285,7 @@
         ]
       },
       "source": [
-        "<font color=\"green\"><b>Naredi sam/a.</b></font> Napovej razrede na testni mno\u017eici. Napovedane razrede primerjaj z resni\u010dnimi in izmeri klasifikacijsko to\u010dnost, natan\u010dnost, priklic in F1-vrednost."
+        "[Odgovor](207-1.ipynb#Odgovor-7-1-2)"
       ]
     },
     {
@@ -1187,36 +2296,7 @@
         ]
       },
       "source": [
-        "<font color=\"green\"><b>Do it yourself.</b></font> Predict the classes on the test set. Compare the predicted classes with the real ones and measure the classification accuracy, precision, recall and F1 value."
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": 14,
-      "metadata": {},
-      "outputs": [
-        {
-          "data": {
-            "text/plain": [
-              "0.771117166212534"
-            ]
-          },
-          "execution_count": 14,
-          "metadata": {},
-          "output_type": "execute_result"
-        }
-      ],
-      "source": [
-        "from sklearn.metrics import accuracy_score\n",
-        "from sklearn.metrics import precision_score\n",
-        "from sklearn.metrics import recall_score\n",
-        "from sklearn.metrics import f1_score\n",
-        "\n",
-        "# uporaba metod: \n",
-        "test_data = pd.read_table('podatki/titanic-test.tab', skiprows=[1,2])\n",
-        "predictions, _ = model.predict(test_data) \n",
-        "truth          = [test_data.loc[i, \"survived\"] for i in range(len(test_data))]\n",
-        "accuracy_score(truth, predictions)"
+        "[Answer](207-1.ipynb#Answer-7-1-2)"
       ]
     },
     {
@@ -1227,7 +2307,9 @@
         ]
       },
       "source": [
-        "<font color=\"orange\"><b>Izziv.</b></font> Nekateri atributi imajo verjetnost 0 pri posameznem razredu. Kako bi popravili klasifikator?"
+        "### Vprašanje 7-1-3\n",
+        "\n",
+        "Poskusite različne [parametre](https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html) pri učenju modela. Kako se spreminja njegova uspešnost?"
       ]
     },
     {
@@ -1238,9 +2320,18 @@
         ]
       },
       "source": [
-        "<font color=\"orange\"><b>Challenge.</b></font> Some attributes have a probability of 0 for each class. How would you repair the classifier?"
+        "### Question 7-1-3\n",
+        "\n",
+        "Try different [parameters](https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html) when training the model. How does its performance change?"
       ]
     },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": []
+    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -1249,7 +2340,7 @@
         ]
       },
       "source": [
-        "<font color=\"blue\"><b>Razmisli.</b></font> Kako bi dopolnili klasifikator, \u010de bi bili nekateri atributi lahko tudi zvezni? Namig: spomni se vaj, ko smo spoznali *verjetnostne porazdelitve* zveznih spremenljivk. "
+        "[Odgovor](207-1.ipynb#Odgovor-7-1-3)"
       ]
     },
     {
@@ -1260,13 +2351,13 @@
         ]
       },
       "source": [
-        "<font color=\"blue\"><b>Think.</b></font> How to complete a classifier if some of the attributes could also be continuous? Hint: remember the exercises when we learned about the probability distributions of continuous variables."
+        "[Answer](207-1.ipynb#Answer-7-1-3)"
       ]
     }
   ],
   "metadata": {
     "kernelspec": {
-      "display_name": "Python 3 (ipykernel)",
+      "display_name": "PRvaje",
       "language": "python",
       "name": "python3"
     },
@@ -1312,4 +2403,4 @@
   },
   "nbformat": 4,
   "nbformat_minor": 1
-}
\ No newline at end of file
+}
diff --git a/notebooks/src/207-1.ipynb b/notebooks/src/207-1.ipynb
index 2dd74b2..874aea4 100644
--- a/notebooks/src/207-1.ipynb
+++ b/notebooks/src/207-1.ipynb
@@ -13,126 +13,665 @@
       "metadata": {},
       "outputs": [],
       "source": [
-        "class NaiveBayes:  \n",
-        "    \"\"\"\n",
-        "    Naive Bayes classifier.\n",
-        "    \n",
-        "    :attribute self.probabilities\n",
-        "        Dictionary that stores\n",
-        "            - prior class probabilities P(Y)\n",
-        "            - attribute probabilities conditional on class P(X|Y)\n",
-        "    \n",
-        "    :attribute self.class_values\n",
-        "        All possible values of the class.\n",
-        "        \n",
-        "    :attribute self.variables\n",
-        "        Variables in the data. \n",
-        "    \n",
-        "    :attribute self.trained\n",
-        "        Set to True after fit is called.\n",
-        "    \"\"\"\n",
-        "    \n",
-        "    def __init__(self):\n",
-        "        self.trained       = False\n",
-        "        self.probabilities = dict()   \n",
-        "    \n",
-        "    \n",
-        "    def fit(self, data):\n",
-        "        \"\"\"\n",
-        "        Fit a NaiveBayes classifier.\n",
-        "        \n",
-        "        :param data\n",
-        "            pandas dataframe.        \n",
-        "        \"\"\"\n",
-        "        class_variable      = data.columns[-1]    # class variable (Y) \n",
-        "        self.class_values   = pd.unique(data['survived'])    # possible class values\n",
-        "        self.variables      = data.columns[:-1]    # all other variables (X)\n",
-        "        \n",
-        "        n = len(data) # number of all data points\n",
-        "        \n",
-        "        # Compute P(Y)\n",
-        "        for y in self.class_values:\n",
+        "import numpy as np\n",
+        "import pandas as pd\n",
+        "from sklearn.tree import DecisionTreeClassifier"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "##### Odgovor 7-1-1"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "##### Answer 7-1-1"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 2,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "nb = np.array([[688., 192.],\n",
+        "       [ 60., 161.]])\n",
+        "majority = np.array([[748., 353.],\n",
+        "       [  0.,   0.]])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 3,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "def vrednotenje(table):\n",
+        "    [[TP,FP,],[FN,TN]] = table\n",
+        "    ca = (TP+TN) / (TP+TN+FP+FN)\n",
+        "    print(\"ca:\\t%.1f\" % (ca*100) + \"%\")\n",
+        "    p = TP / (TP+FP)\n",
+        "    print(\"prec.:\\t%.1f\" % (p*100) + \"%\")\n",
+        "    r = TP / (TP+FN)\n",
+        "    print(\"recall:\\t%.1f\" % (r*100) + \"%\")\n",
+        "    F1 = 2*(p*r)/(p+r)\n",
+        "    print(\"F1:\\t%.1f\" % (F1*100) + \"%\")"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 4,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "ca:\t77.1%\n",
+            "prec.:\t78.2%\n",
+            "recall:\t92.0%\n",
+            "F1:\t84.5%\n"
+          ]
+        }
+      ],
+      "source": [
+        "vrednotenje(nb)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 5,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "ca:\t67.9%\n",
+            "prec.:\t67.9%\n",
+            "recall:\t100.0%\n",
+            "F1:\t80.9%\n"
+          ]
+        }
+      ],
+      "source": [
+        "vrednotenje(majority)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "### Odgovor 7-1-2"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "### Answer 7-1-2"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 6,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/html": [
+              "<style>#sk-container-id-1 {\n",
+              "  /* Definition of color scheme common for light and dark mode */\n",
+              "  --sklearn-color-text: black;\n",
+              "  --sklearn-color-line: gray;\n",
+              "  /* Definition of color scheme for unfitted estimators */\n",
+              "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
+              "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
+              "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
+              "  --sklearn-color-unfitted-level-3: chocolate;\n",
+              "  /* Definition of color scheme for fitted estimators */\n",
+              "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
+              "  --sklearn-color-fitted-level-1: #d4ebff;\n",
+              "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
+              "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
+              "\n",
+              "  /* Specific color for light theme */\n",
+              "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
+              "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
+              "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
+              "  --sklearn-color-icon: #696969;\n",
+              "\n",
+              "  @media (prefers-color-scheme: dark) {\n",
+              "    /* Redefinition of color scheme for dark theme */\n",
+              "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
+              "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
+              "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
+              "    --sklearn-color-icon: #878787;\n",
+              "  }\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 {\n",
+              "  color: var(--sklearn-color-text);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 pre {\n",
+              "  padding: 0;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 input.sk-hidden--visually {\n",
+              "  border: 0;\n",
+              "  clip: rect(1px 1px 1px 1px);\n",
+              "  clip: rect(1px, 1px, 1px, 1px);\n",
+              "  height: 1px;\n",
+              "  margin: -1px;\n",
+              "  overflow: hidden;\n",
+              "  padding: 0;\n",
+              "  position: absolute;\n",
+              "  width: 1px;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-dashed-wrapped {\n",
+              "  border: 1px dashed var(--sklearn-color-line);\n",
+              "  margin: 0 0.4em 0.5em 0.4em;\n",
+              "  box-sizing: border-box;\n",
+              "  padding-bottom: 0.4em;\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-container {\n",
+              "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
+              "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
+              "     so we also need the `!important` here to be able to override the\n",
+              "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
+              "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
+              "  display: inline-block !important;\n",
+              "  position: relative;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-text-repr-fallback {\n",
+              "  display: none;\n",
+              "}\n",
+              "\n",
+              "div.sk-parallel-item,\n",
+              "div.sk-serial,\n",
+              "div.sk-item {\n",
+              "  /* draw centered vertical line to link estimators */\n",
+              "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
+              "  background-size: 2px 100%;\n",
+              "  background-repeat: no-repeat;\n",
+              "  background-position: center center;\n",
+              "}\n",
+              "\n",
+              "/* Parallel-specific style estimator block */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel-item::after {\n",
+              "  content: \"\";\n",
+              "  width: 100%;\n",
+              "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
+              "  flex-grow: 1;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel {\n",
+              "  display: flex;\n",
+              "  align-items: stretch;\n",
+              "  justify-content: center;\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "  position: relative;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel-item {\n",
+              "  display: flex;\n",
+              "  flex-direction: column;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel-item:first-child::after {\n",
+              "  align-self: flex-end;\n",
+              "  width: 50%;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel-item:last-child::after {\n",
+              "  align-self: flex-start;\n",
+              "  width: 50%;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-parallel-item:only-child::after {\n",
+              "  width: 0;\n",
+              "}\n",
+              "\n",
+              "/* Serial-specific style estimator block */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-serial {\n",
+              "  display: flex;\n",
+              "  flex-direction: column;\n",
+              "  align-items: center;\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "  padding-right: 1em;\n",
+              "  padding-left: 1em;\n",
+              "}\n",
+              "\n",
+              "\n",
+              "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
+              "clickable and can be expanded/collapsed.\n",
+              "- Pipeline and ColumnTransformer use this feature and define the default style\n",
+              "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
+              "*/\n",
+              "\n",
+              "/* Pipeline and ColumnTransformer style (default) */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-toggleable {\n",
+              "  /* Default theme specific background. It is overwritten whether we have a\n",
+              "  specific estimator or a Pipeline/ColumnTransformer */\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "}\n",
+              "\n",
+              "/* Toggleable label */\n",
+              "#sk-container-id-1 label.sk-toggleable__label {\n",
+              "  cursor: pointer;\n",
+              "  display: block;\n",
+              "  width: 100%;\n",
+              "  margin-bottom: 0;\n",
+              "  padding: 0.5em;\n",
+              "  box-sizing: border-box;\n",
+              "  text-align: center;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 label.sk-toggleable__label-arrow:before {\n",
+              "  /* Arrow on the left of the label */\n",
+              "  content: \"▸\";\n",
+              "  float: left;\n",
+              "  margin-right: 0.25em;\n",
+              "  color: var(--sklearn-color-icon);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {\n",
+              "  color: var(--sklearn-color-text);\n",
+              "}\n",
+              "\n",
+              "/* Toggleable content - dropdown */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-toggleable__content {\n",
+              "  max-height: 0;\n",
+              "  max-width: 0;\n",
+              "  overflow: hidden;\n",
+              "  text-align: left;\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-toggleable__content.fitted {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-0);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-toggleable__content pre {\n",
+              "  margin: 0.2em;\n",
+              "  border-radius: 0.25em;\n",
+              "  color: var(--sklearn-color-text);\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-toggleable__content.fitted pre {\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-0);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
+              "  /* Expand drop-down */\n",
+              "  max-height: 200px;\n",
+              "  max-width: 100%;\n",
+              "  overflow: auto;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
+              "  content: \"▾\";\n",
+              "}\n",
+              "\n",
+              "/* Pipeline/ColumnTransformer-specific style */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
+              "  color: var(--sklearn-color-text);\n",
+              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
+              "  background-color: var(--sklearn-color-fitted-level-2);\n",
+              "}\n",
+              "\n",
+              "/* Estimator-specific style */\n",
+              "\n",
+              "/* Colorize estimator box */\n",
+              "#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-2);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-label label.sk-toggleable__label,\n",
+              "#sk-container-id-1 div.sk-label label {\n",
+              "  /* The background is the default theme color */\n",
+              "  color: var(--sklearn-color-text-on-default-background);\n",
+              "}\n",
+              "\n",
+              "/* On hover, darken the color of the background */\n",
+              "#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {\n",
+              "  color: var(--sklearn-color-text);\n",
+              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
+              "}\n",
+              "\n",
+              "/* Label box, darken color on hover, fitted */\n",
+              "#sk-container-id-1 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
+              "  color: var(--sklearn-color-text);\n",
+              "  background-color: var(--sklearn-color-fitted-level-2);\n",
+              "}\n",
+              "\n",
+              "/* Estimator label */\n",
+              "\n",
+              "#sk-container-id-1 div.sk-label label {\n",
+              "  font-family: monospace;\n",
+              "  font-weight: bold;\n",
+              "  display: inline-block;\n",
+              "  line-height: 1.2em;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-label-container {\n",
+              "  text-align: center;\n",
+              "}\n",
+              "\n",
+              "/* Estimator-specific */\n",
+              "#sk-container-id-1 div.sk-estimator {\n",
+              "  font-family: monospace;\n",
+              "  border: 1px dotted var(--sklearn-color-border-box);\n",
+              "  border-radius: 0.25em;\n",
+              "  box-sizing: border-box;\n",
+              "  margin-bottom: 0.5em;\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-estimator.fitted {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-0);\n",
+              "}\n",
+              "\n",
+              "/* on hover */\n",
+              "#sk-container-id-1 div.sk-estimator:hover {\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 div.sk-estimator.fitted:hover {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-2);\n",
+              "}\n",
+              "\n",
+              "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
+              "\n",
+              "/* Common style for \"i\" and \"?\" */\n",
+              "\n",
+              ".sk-estimator-doc-link,\n",
+              "a:link.sk-estimator-doc-link,\n",
+              "a:visited.sk-estimator-doc-link {\n",
+              "  float: right;\n",
+              "  font-size: smaller;\n",
+              "  line-height: 1em;\n",
+              "  font-family: monospace;\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "  border-radius: 1em;\n",
+              "  height: 1em;\n",
+              "  width: 1em;\n",
+              "  text-decoration: none !important;\n",
+              "  margin-left: 1ex;\n",
+              "  /* unfitted */\n",
+              "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
+              "  color: var(--sklearn-color-unfitted-level-1);\n",
+              "}\n",
+              "\n",
+              ".sk-estimator-doc-link.fitted,\n",
+              "a:link.sk-estimator-doc-link.fitted,\n",
+              "a:visited.sk-estimator-doc-link.fitted {\n",
+              "  /* fitted */\n",
+              "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
+              "  color: var(--sklearn-color-fitted-level-1);\n",
+              "}\n",
+              "\n",
+              "/* On hover */\n",
+              "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
+              ".sk-estimator-doc-link:hover,\n",
+              "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
+              ".sk-estimator-doc-link:hover {\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-3);\n",
+              "  color: var(--sklearn-color-background);\n",
+              "  text-decoration: none;\n",
+              "}\n",
+              "\n",
+              "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
+              ".sk-estimator-doc-link.fitted:hover,\n",
+              "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
+              ".sk-estimator-doc-link.fitted:hover {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-3);\n",
+              "  color: var(--sklearn-color-background);\n",
+              "  text-decoration: none;\n",
+              "}\n",
+              "\n",
+              "/* Span, style for the box shown on hovering the info icon */\n",
+              ".sk-estimator-doc-link span {\n",
+              "  display: none;\n",
+              "  z-index: 9999;\n",
+              "  position: relative;\n",
+              "  font-weight: normal;\n",
+              "  right: .2ex;\n",
+              "  padding: .5ex;\n",
+              "  margin: .5ex;\n",
+              "  width: min-content;\n",
+              "  min-width: 20ex;\n",
+              "  max-width: 50ex;\n",
+              "  color: var(--sklearn-color-text);\n",
+              "  box-shadow: 2pt 2pt 4pt #999;\n",
+              "  /* unfitted */\n",
+              "  background: var(--sklearn-color-unfitted-level-0);\n",
+              "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
+              "}\n",
+              "\n",
+              ".sk-estimator-doc-link.fitted span {\n",
+              "  /* fitted */\n",
+              "  background: var(--sklearn-color-fitted-level-0);\n",
+              "  border: var(--sklearn-color-fitted-level-3);\n",
+              "}\n",
+              "\n",
+              ".sk-estimator-doc-link:hover span {\n",
+              "  display: block;\n",
+              "}\n",
+              "\n",
+              "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
+              "\n",
+              "#sk-container-id-1 a.estimator_doc_link {\n",
+              "  float: right;\n",
+              "  font-size: 1rem;\n",
+              "  line-height: 1em;\n",
+              "  font-family: monospace;\n",
+              "  background-color: var(--sklearn-color-background);\n",
+              "  border-radius: 1rem;\n",
+              "  height: 1rem;\n",
+              "  width: 1rem;\n",
+              "  text-decoration: none;\n",
+              "  /* unfitted */\n",
+              "  color: var(--sklearn-color-unfitted-level-1);\n",
+              "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 a.estimator_doc_link.fitted {\n",
+              "  /* fitted */\n",
+              "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
+              "  color: var(--sklearn-color-fitted-level-1);\n",
+              "}\n",
+              "\n",
+              "/* On hover */\n",
+              "#sk-container-id-1 a.estimator_doc_link:hover {\n",
+              "  /* unfitted */\n",
+              "  background-color: var(--sklearn-color-unfitted-level-3);\n",
+              "  color: var(--sklearn-color-background);\n",
+              "  text-decoration: none;\n",
+              "}\n",
+              "\n",
+              "#sk-container-id-1 a.estimator_doc_link.fitted:hover {\n",
+              "  /* fitted */\n",
+              "  background-color: var(--sklearn-color-fitted-level-3);\n",
+              "}\n",
+              "</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>DecisionTreeClassifier(criterion=&#x27;entropy&#x27;, max_depth=4, random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;DecisionTreeClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.4/modules/generated/sklearn.tree.DecisionTreeClassifier.html\">?<span>Documentation for DecisionTreeClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>DecisionTreeClassifier(criterion=&#x27;entropy&#x27;, max_depth=4, random_state=42)</pre></div> </div></div></div></div>"
+            ],
+            "text/plain": [
+              "DecisionTreeClassifier(criterion='entropy', max_depth=4, random_state=42)"
+            ]
+          },
+          "execution_count": 6,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "data = pd.read_table('../data/titanic-training.tab', skiprows=[1,2])\n",
+        "data[\"status\"] = data[\"status\"].astype(\"category\").cat.codes\n",
+        "data[\"age\"] = data[\"age\"].astype(\"category\").cat.codes   # child=1, adult=0\n",
+        "data[\"sex\"] = data[\"sex\"].astype(\"category\").cat.codes   # male=1, female=0\n",
+        "\n",
+        "X = data[[\"status\", \"age\", \"sex\"]]\n",
+        "y = data[\"survived\"]\n",
+        "\n",
+        "tree = DecisionTreeClassifier(\n",
+        "    criterion=\"entropy\",\n",
+        "    max_depth=4,\n",
+        "    random_state=42\n",
+        ")\n",
+        "tree.fit(X, y)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 7,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "test_data = pd.read_table('../data/titanic-test.tab', skiprows=[1,2])\n",
+        "test_data[\"status\"] = test_data[\"status\"].astype(\"category\").cat.codes\n",
+        "test_data[\"age\"] = test_data[\"age\"].astype(\"category\").cat.codes   # child=1, adult=0\n",
+        "test_data[\"sex\"] = test_data[\"sex\"].astype(\"category\").cat.codes   # male=1, female=0\n",
+        "predicts = tree.predict(test_data[[\"status\", \"age\", \"sex\"]])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 8,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "ca:\t78.3%\n",
+            "prec.:\t94.5%\n",
+            "recall:\t34.3%\n",
+            "F1:\t50.3%\n"
+          ]
+        }
+      ],
+      "source": [
+        "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n",
+        "\n",
+        "ca = accuracy_score(test_data[\"survived\"], predicts)\n",
+        "p = precision_score(test_data[\"survived\"], predicts, pos_label='yes')\n",
+        "r = recall_score(test_data[\"survived\"], predicts, pos_label='yes')\n",
+        "F1 = f1_score(test_data[\"survived\"], predicts, pos_label='yes')\n",
+        "\n",
+        "print(\"ca:\\t%.1f\" % (ca*100) + \"%\")\n",
+        "print(\"prec.:\\t%.1f\" % (p*100) + \"%\")\n",
+        "print(\"recall:\\t%.1f\" % (r*100) + \"%\")\n",
+        "print(\"F1:\\t%.1f\" % (F1*100) + \"%\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "sl"
+        ]
+      },
+      "source": [
+        "### Odgovor 7-1-3"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "tags": [
+          "en"
+        ]
+      },
+      "source": [
+        "### Answer 7-1-3"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 9,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "ca:\t77.7%\n",
+            "prec.:\t94.2%\n",
+            "recall:\t32.3%\n",
+            "F1:\t48.1%\n"
+          ]
+        }
+      ],
+      "source": [
+        "tree = DecisionTreeClassifier(\n",
+        "    criterion=\"gini\",\n",
+        "    splitter=\"random\",\n",
+        "    max_depth=2,\n",
+        "    min_samples_split=10,\n",
+        "    min_samples_leaf=5,\n",
+        "    random_state=42\n",
+        ")\n",
+        "tree.fit(X, y)\n",
+        "\n",
+        "predicts = tree.predict(test_data[[\"status\", \"age\", \"sex\"]])\n",
+        "ca = accuracy_score(test_data[\"survived\"], predicts)\n",
+        "p = precision_score(test_data[\"survived\"], predicts, pos_label='yes')\n",
+        "r = recall_score(test_data[\"survived\"], predicts, pos_label='yes')\n",
+        "F1 = f1_score(test_data[\"survived\"], predicts, pos_label='yes')\n",
         "\n",
-        "            # Compute class probabilities and correctly fill\n",
-        "            #   probabilities[y] = ... \n",
-        "            # Select all examples (rows) with class = y\n",
-        "            data_subset = data.loc[data[class_variable] == y]\n",
-        "            m = len(data_subset)\n",
-        "        \n",
-        "            self.probabilities[y] = m/n\n",
-        "        \n",
-        "        # Compute P(X|Y)\n",
-        "        for y in self.class_values:\n",
-        "            \n",
-        "            # Select all examples (rows) with class = y\n",
-        "            data_subset = data.loc[data[class_variable] == y]\n",
-        "            p = len(data_subset)\n",
-        "            \n",
-        "            for variable in self.variables:\n",
-        "                for x in pd.unique(data[variable]):\n",
-        "                    \n",
-        "                    # Compute correct conditional class probability\n",
-        "                    #   probabilities[x, value, c] = ... \n",
-        "                    # \n",
-        "                    # Select all examples with class == y AND \n",
-        "                    # variable x == value\n",
-        "                    # Hint: use SameValue filter twice\n",
-        "                    data_subset = data.loc[(data[variable] == x) & (data[class_variable] == y)]\n",
-        "                    m = len(data_subset)\n",
-        "                    \n",
-        "                    self.probabilities[variable, x, y] = m/p\n",
-        "    \n",
-        "        self.trained = True\n",
-        "           \n",
-        "    def predict_instance(self, row):\n",
-        "        \"\"\"\n",
-        "        Predict a class value for one row.\n",
-        "        \n",
-        "        :param row\n",
-        "            Orange data Instance.\n",
-        "        :return \n",
-        "            Class prediction.\n",
-        "        \"\"\"\n",
-        "        curr_p = float(\"-inf\")   # Current highest \"probability\" (unnormalized)\n",
-        "        curr_c = None            # Current most probable class\n",
-        "        \n",
-        "        for y in self.class_values:\n",
-        "            p = np.log(self.probabilities[y])\n",
-        "            for x in self.variables:\n",
-        "                p = p + np.log(self.probabilities[x, row[x], y])\n",
-        "            \n",
-        "            if p > curr_p:\n",
-        "                curr_p = p\n",
-        "                curr_c = y\n",
-        "                \n",
-        "        return curr_c, curr_p\n",
-        "        \n",
-        "    def predict(self, data):\n",
-        "        \"\"\"\n",
-        "        Predict class labels for all rows in data.\n",
-        "        \n",
-        "        :param data\n",
-        "            Orange data Table.       \n",
-        "        :return y\n",
-        "            NumPy vector with predicted classes.\n",
-        "        \"\"\"\n",
-        "        \n",
-        "        n = len(data)\n",
-        "        predictions = list()\n",
-        "        confidences = np.zeros((n, ))\n",
-        "        \n",
-        "        for i in range(len(data)):\n",
-        "            pred, cf = self.predict_instance(data.iloc[i])\n",
-        "            predictions.append(pred)\n",
-        "            confidences[i] = cf\n",
-        "    \n",
-        "        return predictions, confidences"
+        "print(\"ca:\\t%.1f\" % (ca*100) + \"%\")\n",
+        "print(\"prec.:\\t%.1f\" % (p*100) + \"%\")\n",
+        "print(\"recall:\\t%.1f\" % (r*100) + \"%\")\n",
+        "print(\"F1:\\t%.1f\" % (F1*100) + \"%\")"
       ]
     }
   ],
   "metadata": {
     "kernelspec": {
-      "display_name": "Python 3 (ipykernel)",
+      "display_name": "PRvaje",
       "language": "python",
       "name": "python3"
     },