fixed distribution samples in examples

Author: awarebayes

docs/source/examples/cloning_from_github.rst

@@ -0,0 +1,55 @@
+Cloning from github
+===================
+
+Pro tip: clone without history (unless you need it)::
+
+    git clone --depth 1 [email protected]:awarebayes/RecNN.git 
+
+Create ENV and install deps::
+
+    conda create --name recnn
+    conda activate recnn
+    cd RecNN
+    pip install -r requirements.txt
+
+Download data from the donwloads section
+
+Start jupyter notebook and jump to the examples folder ::
+
+    jupyter-notebook .
+
+Here is how my project directories looks like (shallow)::
+
+    RecNN
+    ├── .circleci
+    ├── data
+    ├── docs
+    ├── examples
+    ├── .git
+    ├── .gitignore
+    ├── LICENSE
+    ├── models
+    ├── readme.md
+    ├── recnn
+    ├── requirements.txt
+    ├── res
+    ├── runs
+    ├── setup.cfg
+    └── setup.py
+
+Here is the data directory (ignore the cache)::
+
+    data
+    ├── cache
+    │   ├── frame_env.pkl
+    │   └── frame_env_truncated.pkl
+    ├── embeddings
+    │   └── ml20_pca128.pkl
+    └── ml-20m
+        ├── genome-scores.csv
+        ├── genome-tags.csv
+        ├── links.csv
+        ├── movies.csv
+        ├── ratings.csv
+        ├── README.txt
+        └── tags.csv

examples/1. Vanilla RL/2. DDPG.ipynb

@@ -230,7 +230,7 @@
     "    losses = ddpg_update(test_batch, params, learn=False, step=step)\n",
     "    \n",
     "    gen_actions = debug['next_action']\n",
-    "    true_actions = env.embeddings.detach().cpu().numpy()\n",
+    "    true_actions = env.base.embeddings.detach().cpu().numpy()\n",
     "    \n",
     "    f = plotter.kde_reconstruction_error(ad, gen_actions, true_actions, cuda)\n",
     "    writer.add_figure('rec_error',f, losses['step'])\n",
@@ -488,7 +488,7 @@
    ],
    "source": [
     "gen_actions = debug['next_action']\n",
-    "true_actions = env.embeddings.numpy()\n",
+    "true_actions = env.base.embeddings.numpy()\n",
     "\n",
     "\n",
     "ad = recnn.nn.AnomalyDetector().to(cuda)\n",

examples/1. Vanilla RL/3. TD3.ipynb

@@ -23,7 +23,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "import torch\n",
+    "import torchconda install faiss-gpu cudatoolkit=10.0 -c pytorch # For CUDA10\n",
+    "\n",
     "from torch.utils.data import Dataset, DataLoader\n",
     "import torch.nn as nn\n",
     "import torch.nn.functional as F\n",
@@ -37,12 +38,10 @@
    "outputs": [],
    "source": [
     "import numpy as np\n",
-    "from scipy.spatial import distance\n",
     "from tqdm.auto import tqdm\n",
     "import pickle\n",
     "import gc\n",
     "import json\n",
-    "import h5py\n",
     "import pandas as pd\n",
     "\n",
     "from IPython.display import clear_output\n",
@@ -70,14 +69,12 @@
    "metadata": {},
    "outputs": [
     {
+     "output_type": "execute_result",
      "data": {
-      "text/plain": [
-       "<All keys matched successfully>"
-      ]
+      "text/plain": "<All keys matched successfully>"
      },
-     "execution_count": 3,
      "metadata": {},
-     "output_type": "execute_result"
+     "execution_count": 3
     }
    ],
    "source": [
@@ -91,105 +88,31 @@
    "cell_type": "code",
    "execution_count": 4,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "5d711e3a8edd4108a6bdd94a47712092",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "HBox(children=(IntProgress(value=0, max=20000263), HTML(value='')))"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "3ea67abab15a4e08bb0b13e7b342e34d",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "HBox(children=(IntProgress(value=0, max=20000263), HTML(value='')))"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "d8120ad7d9724cb9b1c883d80ba2c24a",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "HBox(children=(IntProgress(value=0, max=138493), HTML(value='')))"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
+    "frame_size = 10\n",
+    "batch_size = 1\n",
     "# embeddgings: https://drive.google.com/open?id=1EQ_zXBR3DKpmJR3jBgLvt-xoOvArGMsL\n",
-    "env = recnn.data.FrameEnv('../../data/embeddings/ml20_pca128.pkl',\n",
-    "                         '../../data/ml-20m/ratings.csv', 10, 1)"
+    "dirs = recnn.data.env.DataPath(\n",
+    "    base=\"../../data/\",\n",
+    "    embeddings=\"embeddings/ml20_pca128.pkl\",\n",
+    "    ratings=\"ml-20m/ratings.csv\",\n",
+    "    cache=\"cache/frame_env.pkl\", # cache will generate after you run\n",
+    "    use_cache=True\n",
+    ")\n",
+    "env = recnn.data.env.FrameEnv(dirs, frame_size, batch_size)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
     "test_batch = next(iter(env.test_dataloader))\n",
     "state, action, reward, next_state, done = recnn.data.get_base_batch(test_batch)"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def rank(gen_action, metric):\n",
-    "    scores = []\n",
-    "    for i in movie_embeddings_key_dict.keys():\n",
-    "        scores.append([i, metric(movie_embeddings_key_dict[i], gen_action)])\n",
-    "    scores = list(sorted(scores, key = lambda x: x[1]))\n",
-    "    scores = scores[:10]\n",
-    "    ids = [i[0] for i in scores]\n",
-    "    dist = [i[1] for i in scores]\n",
-    "\n",
-    "    return ids, dist"
-   ]
-  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -199,15 +122,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/home/dev/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:7: FutureWarning: arrays to stack must be passed as a \"sequence\" type such as list or tuple. Support for non-sequence iterables such as generators is deprecated as of NumPy 1.16 and will raise an error in the future.\n",
-      "  import sys\n"
+     "output_type": "error",
+     "ename": "ModuleNotFoundError",
+     "evalue": "No module named 'faiss'",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
+      "\u001b[0;32m<ipython-input-7-f4db3fa501af>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mimport\u001b[0m \u001b[0mfaiss\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0;31m# test indexes\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mindexL2\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfaiss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mIndexFlatL2\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m128\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mindexIP\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfaiss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mIndexFlatIP\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m128\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mindexCOS\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfaiss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mIndexFlatIP\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m128\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'faiss'"
      ]
     }
    ],
@@ -218,7 +144,7 @@
     "indexIP = faiss.IndexFlatIP(128)\n",
     "indexCOS = faiss.IndexFlatIP(128)\n",
     "\n",
-    "mov_mat  = np.stack(env.movie_embeddings_key_dict.values()).astype('float32')\n",
+    "mov_mat  = env.base.embeddings.detach().cpu().numpy().astype('float32')\n",
     "indexL2.add(mov_mat)\n",
     "indexIP.add(mov_mat)\n",
     "indexCOS.add(normalize(mov_mat, axis=1, norm='l2'))\n",
@@ -1247,9 +1173,9 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.3"
+   "version": "3.8.5-final"
   }
  },
  "nbformat": 4,
  "nbformat_minor": 2
-}
+}