diff --git a/Makefile b/Makefile
index 5e88121d..56751110 100644
--- a/Makefile
+++ b/Makefile
@@ -25,4 +25,4 @@ actions = \
 
 .PHONY: $(actions)
 $(actions):
-	@python scripts/make "$@"
+	@python3 scripts/make "$@"
diff --git a/README.md b/README.md
index 7c2e754a..b0d2a1c5 100644
--- a/README.md
+++ b/README.md
@@ -5,6 +5,12 @@
 [![pypi version](https://img.shields.io/pypi/v/stimulus-py.svg)](https://pypi.org/project/stimulus-py/)
 [![gitter](https://badges.gitter.im/join%20chat.svg)](https://app.gitter.im/#/room/#stimulus-py:gitter.im)
 
+## Introduction
+
+Most (if not all) quality software is thouroughly tested. Deep neural networks seem to have escaped this paradigm. 
+
+Here, we attempt at solving the testing problem by proposing an extensive library to test deep neural networks beyond test-set performance. 
+
 
 
 ## Installation
diff --git a/pyproject.toml b/pyproject.toml
index f9c17b36..e5dd4d68 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -42,6 +42,7 @@ dependencies = [
     "scikit-learn>=1.5.0",
     "scipy==1.14.1",
     "torch>=2.2.2",
+    "torch==2.2.2; sys_platform == 'darwin' and platform_machine == 'x86_64'"
 ]
 
 [project.scripts]