diff --git a/examples/colab_MovieLen1M_YoutubeDNN.ipynb b/examples/colab_MovieLen1M_YoutubeDNN.ipynb index 57decf2..27e2654 100644 --- a/examples/colab_MovieLen1M_YoutubeDNN.ipynb +++ b/examples/colab_MovieLen1M_YoutubeDNN.ipynb @@ -22,7 +22,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 1, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -30,14 +30,49 @@ "id": "yTl6d6jO1oqf", "outputId": "ee7303f1-8970-4726-a9f1-368798077228" }, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "--2024-05-13 17:38:02-- http://files.grouplens.org/datasets/movielens/ml-1m.zip\n", + "正在查找主機 files.grouplens.org (files.grouplens.org)... 128.101.65.152\n", + "正在連接 files.grouplens.org (files.grouplens.org)|128.101.65.152|:80... 連上了。\n", + "已送出 HTTP 要求，正在等候回應... 200 OK\n", + "長度: 5917549 (5.6M) [application/zip]\n", + "儲存到：「./ml-1m.zip」\n", + "\n", + "./ml-1m.zip 100%[===================>] 5.64M 1.41MB/s 於 4.7s \n", + "\n", + "2024-05-13 17:38:08 (1.20 MB/s) - 已儲存 「./ml-1m.zip」 [5917549/5917549]\n", + "\n", + "--2024-05-13 17:38:08-- https://raw.githubusercontent.com/shenweichen/DeepMatch/master/examples/preprocess.py\n", + "正在查找主機 raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.109.133, 185.199.110.133, 185.199.111.133, ...\n", + "正在連接 raw.githubusercontent.com (raw.githubusercontent.com)|185.199.109.133|:443... 連上了。\n", + "已送出 HTTP 要求，正在等候回應... 200 OK\n", + "長度: 6705 (6.5K) [text/plain]\n", + "儲存到：「preprocess.py」\n", + "\n", + "preprocess.py 100%[===================>] 6.55K --.-KB/s 於 0s \n", + "\n", + "2024-05-13 17:38:08 (22.6 MB/s) - 已儲存 「preprocess.py」 [6705/6705]\n", + "\n", + "Archive: ml-1m.zip\n", + " creating: ml-1m/\n", + " inflating: ml-1m/movies.dat \n", + " inflating: ml-1m/ratings.dat \n", + " inflating: ml-1m/README \n", + " inflating: ml-1m/users.dat \n" + ] + } + ], "source": [ "! wget http://files.grouplens.org/datasets/movielens/ml-1m.zip -O ./ml-1m.zip \n", "! wget https://raw.githubusercontent.com/shenweichen/DeepMatch/master/examples/preprocess.py -O preprocess.py\n", "! unzip -o ml-1m.zip \n", - "! pip uninstall -y -q tensorflow\n", - "! pip install -q tensorflow-gpu==2.5.0\n", - "! pip install -q deepmatch" + "# ! pip uninstall -y -q tensorflow\n", + "# ! pip install -q tensorflow-gpu==2.5.0\n", + "# ! pip install -q deepmatch" ] }, { @@ -68,6 +103,15 @@ "from deepmatch.utils import sampledsoftmaxloss, NegativeSampler" ] }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [], + "source": [ + "from IPython.display import display" + ] + }, { "cell_type": "markdown", "metadata": { @@ -79,7 +123,37 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[1m\u001b[36m__pycache__\u001b[m\u001b[m\n", + "colab_MovieLen1M_ComiRec.ipynb\n", + "colab_MovieLen1M_DSSM_InBatchSoftmax.ipynb\n", + "colab_MovieLen1M_SDM.ipynb\n", + "colab_MovieLen1M_YoutubeDNN.ipynb\n", + "\u001b[1m\u001b[36mml-1m\u001b[m\u001b[m\n", + "ml-1m.zip\n", + "movielens_sample.txt\n", + "preprocess.py\n", + "run_dssm_inbatchsoftmax.py\n", + "run_dssm_negsampling.py\n", + "run_ncf.py\n", + "run_sdm.py\n", + "run_youtubednn.py\n" + ] + } + ], + "source": [ + "!ls" + ] + }, + { + "cell_type": "code", + "execution_count": 4, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -92,12 +166,12 @@ "name": "stderr", "output_type": "stream", "text": [ - "/Users/swc/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:4: ParserWarning: Falling back to the 'python' engine because the 'c' engine does not support regex separators (separators > 1 char and different from '\\s+' are interpreted as regex); you can avoid this warning by specifying engine='python'.\n", - " after removing the cwd from sys.path.\n", - "/Users/swc/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:6: ParserWarning: Falling back to the 'python' engine because the 'c' engine does not support regex separators (separators > 1 char and different from '\\s+' are interpreted as regex); you can avoid this warning by specifying engine='python'.\n", - " \n", - "/Users/swc/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:8: ParserWarning: Falling back to the 'python' engine because the 'c' engine does not support regex separators (separators > 1 char and different from '\\s+' are interpreted as regex); you can avoid this warning by specifying engine='python'.\n", - " \n" + "/var/folders/zk/7_lr6y6s5wx2pcwhkjyfln3m0000gn/T/ipykernel_34583/1533986894.py:4: ParserWarning: Falling back to the 'python' engine because the 'c' engine does not support regex separators (separators > 1 char and different from '\\s+' are interpreted as regex); you can avoid this warning by specifying engine='python'.\n", + " user = pd.read_csv(data_path+'ml-1m/users.dat',sep='::',header=None,names=unames)\n", + "/var/folders/zk/7_lr6y6s5wx2pcwhkjyfln3m0000gn/T/ipykernel_34583/1533986894.py:6: ParserWarning: Falling back to the 'python' engine because the 'c' engine does not support regex separators (separators > 1 char and different from '\\s+' are interpreted as regex); you can avoid this warning by specifying engine='python'.\n", + " ratings = pd.read_csv(data_path+'ml-1m/ratings.dat',sep='::',header=None,names=rnames)\n", + "/var/folders/zk/7_lr6y6s5wx2pcwhkjyfln3m0000gn/T/ipykernel_34583/1533986894.py:8: ParserWarning: Falling back to the 'python' engine because the 'c' engine does not support regex separators (separators > 1 char and different from '\\s+' are interpreted as regex); you can avoid this warning by specifying engine='python'.\n", + " movies = pd.read_csv(data_path+'ml-1m/movies.dat',sep='::',header=None,names=mnames,encoding=\"unicode_escape\")\n" ] } ], @@ -112,9 +186,434 @@ "movies = pd.read_csv(data_path+'ml-1m/movies.dat',sep='::',header=None,names=mnames,encoding=\"unicode_escape\")\n", "movies['genres'] = list(map(lambda x: x.split('|')[0], movies['genres'].values))\n", "\n", - "data = pd.merge(pd.merge(ratings,movies),user)#.iloc[:10000]" + "data = pd.merge(pd.merge(ratings,movies),user).iloc[:10000]\n", + "# data = pd.merge(pd.merge(ratings,movies),user)" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(10000, 10)" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
user_idmovie_idratingtimestamptitlegenresgenderageoccupationzip
0111935978300760One Flew Over the Cuckoo's Nest (1975)DramaF11048067
116613978302109James and the Giant Peach (1996)AnimationF11048067
219143978301968My Fair Lady (1964)MusicalF11048067
3134084978300275Erin Brockovich (2000)DramaF11048067
4123555978824291Bug's Life, A (1998)AnimationF11048067
\n", + "
" + ], + "text/plain": [ + " user_id movie_id rating timestamp \\\n", + "0 1 1193 5 978300760 \n", + "1 1 661 3 978302109 \n", + "2 1 914 3 978301968 \n", + "3 1 3408 4 978300275 \n", + "4 1 2355 5 978824291 \n", + "\n", + " title genres gender age occupation \\\n", + "0 One Flew Over the Cuckoo's Nest (1975) Drama F 1 10 \n", + "1 James and the Giant Peach (1996) Animation F 1 10 \n", + "2 My Fair Lady (1964) Musical F 1 10 \n", + "3 Erin Brockovich (2000) Drama F 1 10 \n", + "4 Bug's Life, A (1998) Animation F 1 10 \n", + "\n", + " zip \n", + "0 48067 \n", + "1 48067 \n", + "2 48067 \n", + "3 48067 \n", + "4 48067 " + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
user_idmovie_idratingtimestamptitlegenresgenderageoccupationzip
count10000.00000010000.00000010000.0000001.000000e+0410000100001000010000.00000010000.00000010000
uniqueNaNNaNNaNNaN2175182NaNNaN46
topNaNNaNNaNNaNOne Flew Over the Cuckoo's Nest (1975)ComedyMNaNNaN96931
freqNaNNaNNaNNaN4625366866NaNNaN684
mean91.3629001840.5577003.6479009.807671e+08NaNNaNNaN28.9363007.295300NaN
std54.2528041075.3772341.0527281.120014e+07NaNNaNNaN11.5535366.484219NaN
min1.0000001.0000001.0000009.770896e+08NaNNaNNaN1.0000000.000000NaN
25%48.0000001030.7500003.0000009.774469e+08NaNNaNNaN25.0000002.000000NaN
50%81.0000001732.5000004.0000009.779737e+08NaNNaNNaN25.0000004.000000NaN
75%146.0000002722.0000004.0000009.781587e+08NaNNaNNaN35.00000010.000000NaN
max180.0000003952.0000005.0000001.043799e+09NaNNaNNaN56.00000020.000000NaN
\n", + "
" + ], + "text/plain": [ + " user_id movie_id rating timestamp \\\n", + "count 10000.000000 10000.000000 10000.000000 1.000000e+04 \n", + "unique NaN NaN NaN NaN \n", + "top NaN NaN NaN NaN \n", + "freq NaN NaN NaN NaN \n", + "mean 91.362900 1840.557700 3.647900 9.807671e+08 \n", + "std 54.252804 1075.377234 1.052728 1.120014e+07 \n", + "min 1.000000 1.000000 1.000000 9.770896e+08 \n", + "25% 48.000000 1030.750000 3.000000 9.774469e+08 \n", + "50% 81.000000 1732.500000 4.000000 9.779737e+08 \n", + "75% 146.000000 2722.000000 4.000000 9.781587e+08 \n", + "max 180.000000 3952.000000 5.000000 1.043799e+09 \n", + "\n", + " title genres gender age \\\n", + "count 10000 10000 10000 10000.000000 \n", + "unique 2175 18 2 NaN \n", + "top One Flew Over the Cuckoo's Nest (1975) Comedy M NaN \n", + "freq 46 2536 6866 NaN \n", + "mean NaN NaN NaN 28.936300 \n", + "std NaN NaN NaN 11.553536 \n", + "min NaN NaN NaN 1.000000 \n", + "25% NaN NaN NaN 25.000000 \n", + "50% NaN NaN NaN 25.000000 \n", + "75% NaN NaN NaN 35.000000 \n", + "max NaN NaN NaN 56.000000 \n", + "\n", + " occupation zip \n", + "count 10000.000000 10000 \n", + "unique NaN 46 \n", + "top NaN 96931 \n", + "freq NaN 684 \n", + "mean 7.295300 NaN \n", + "std 6.484219 NaN \n", + "min 0.000000 NaN \n", + "25% 2.000000 NaN \n", + "50% 4.000000 NaN \n", + "75% 10.000000 NaN \n", + "max 20.000000 NaN " + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'movie_ids : '" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "2175" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'user_ids : '" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "46" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "display(\n", + " data.shape,\n", + " data.head(),\n", + " data.describe(include='all'),\n", + " 'movie_ids : ',\n", + " data.movie_id.nunique(),\n", + " 'user_ids : ', \n", + " data.user_id.nunique()\n", + ")" ] }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, { "cell_type": "markdown", "metadata": { @@ -126,7 +625,67 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "100%|█████████████████████████████████████████████████████████████████████████████████| 46/46 [00:00<00:00, 552.40it/s]\n", + "/Users/yulongtsai/Desktop/Working_Area/DeepMatch/examples/preprocess.py:86: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n", + " train_seq_genres = np.array([line[5] for line in train_set])\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "8 8\n" + ] + } + ], + "source": [ + "#data = pd.read_csvdata = pd.read_csv(\"./movielens_sample.txt\")\n", + "sparse_features = [\"movie_id\", \"user_id\",\n", + " \"gender\", \"age\", \"occupation\", \"zip\", \"genres\"]\n", + "SEQ_LEN = 50\n", + "negsample = 0\n", + "\n", + "# 1.Label Encoding for sparse features,and process sequence features with `gen_date_set` and `gen_model_input`\n", + "\n", + "feature_max_idx = {}\n", + "for feature in sparse_features:\n", + " lbe = LabelEncoder()\n", + " data[feature] = lbe.fit_transform(data[feature]) + 1\n", + " feature_max_idx[feature] = data[feature].max() + 1\n", + "\n", + "user_profile = data[[\"user_id\", \"gender\", \"age\", \"occupation\", \"zip\"]].drop_duplicates('user_id')\n", + "\n", + "item_profile = data[[\"movie_id\"]].drop_duplicates('movie_id')\n", + "\n", + "user_profile.set_index(\"user_id\", inplace=True)\n", + "\n", + "user_item_list = data.groupby(\"user_id\")['movie_id'].apply(list)\n", + "\n", + "train_set, test_set = gen_data_set(data, SEQ_LEN, negsample)\n", + "\n", + "train_model_input, train_label = gen_model_input(train_set, user_profile, SEQ_LEN)\n", + "test_model_input, test_label = gen_model_input(test_set, user_profile, SEQ_LEN)" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": {}, + "outputs": [], + "source": [ + "# sampler_config" + ] + }, + { + "cell_type": "code", + "execution_count": 7, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -135,94 +694,80 @@ "outputId": "962afe1c-d387-4345-861f-e9b974a0b495" }, "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Train on 9908 samples\n", + "Epoch 1/20\n" + ] + }, { "name": "stderr", "output_type": "stream", "text": [ - "100%|██████████| 6040/6040 [00:12<00:00, 488.35it/s]\n" + "2024-05-14 11:57:47.790847: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA\n", + "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "8 8\n", - "Train on 988129 samples\n", - "Epoch 1/20\n", - "988129/988129 [==============================] - 38s 39us/sample - loss: 5.6344\n", + "9908/9908 [==============================] - 1s 66us/sample - loss: 11.8385\n", "Epoch 2/20\n", - "988129/988129 [==============================] - 41s 41us/sample - loss: 4.6947\n", + "9908/9908 [==============================] - 0s 27us/sample - loss: 9.2014\n", "Epoch 3/20\n", - "988129/988129 [==============================] - 39s 39us/sample - loss: 4.4681\n", + "9908/9908 [==============================] - 0s 29us/sample - loss: 8.7334\n", "Epoch 4/20\n", - "988129/988129 [==============================] - 38s 38us/sample - loss: 4.3227\n", + "9908/9908 [==============================] - 0s 26us/sample - loss: 8.3460\n", "Epoch 5/20\n", - "988129/988129 [==============================] - 38s 38us/sample - loss: 4.2224\n", + "9908/9908 [==============================] - 0s 27us/sample - loss: 7.8493\n", "Epoch 6/20\n", - "988129/988129 [==============================] - 37s 37us/sample - loss: 4.1463\n", + "9908/9908 [==============================] - 0s 27us/sample - loss: 7.4769\n", "Epoch 7/20\n", - "988129/988129 [==============================] - 37s 37us/sample - loss: 4.0843\n", + "9908/9908 [==============================] - 0s 27us/sample - loss: 7.1053\n", "Epoch 8/20\n", - "988129/988129 [==============================] - 37s 38us/sample - loss: 4.0339\n", + "9908/9908 [==============================] - 0s 28us/sample - loss: 6.9057\n", "Epoch 9/20\n", - "988129/988129 [==============================] - 44s 44us/sample - loss: 3.9941\n", + "9908/9908 [==============================] - 0s 27us/sample - loss: 6.4999\n", "Epoch 10/20\n", - "988129/988129 [==============================] - 38s 38us/sample - loss: 3.9619\n", + "9908/9908 [==============================] - 0s 27us/sample - loss: 6.2812\n", "Epoch 11/20\n", - "988129/988129 [==============================] - 43s 43us/sample - loss: 3.9349\n", + "9908/9908 [==============================] - 0s 28us/sample - loss: 6.0093\n", "Epoch 12/20\n", - "988129/988129 [==============================] - 39s 39us/sample - loss: 3.9112\n", + "9908/9908 [==============================] - 0s 28us/sample - loss: 5.7788\n", "Epoch 13/20\n", - "988129/988129 [==============================] - 39s 39us/sample - loss: 3.8902\n", + "9908/9908 [==============================] - 0s 27us/sample - loss: 5.6211\n", "Epoch 14/20\n", - "988129/988129 [==============================] - 39s 39us/sample - loss: 3.8712\n", + "9908/9908 [==============================] - 0s 28us/sample - loss: 5.4398\n", "Epoch 15/20\n", - "988129/988129 [==============================] - 38s 38us/sample - loss: 3.8560\n", + "9908/9908 [==============================] - 0s 29us/sample - loss: 5.3278\n", "Epoch 16/20\n", - "988129/988129 [==============================] - 39s 40us/sample - loss: 3.8413\n", + "9908/9908 [==============================] - 0s 35us/sample - loss: 5.2360\n", "Epoch 17/20\n", - "988129/988129 [==============================] - 39s 39us/sample - loss: 3.8285\n", + "9908/9908 [==============================] - 0s 32us/sample - loss: 5.1163\n", "Epoch 18/20\n", - "988129/988129 [==============================] - 38s 38us/sample - loss: 3.8185\n", + "9908/9908 [==============================] - 0s 34us/sample - loss: 4.9778\n", "Epoch 19/20\n", - "988129/988129 [==============================] - 40s 40us/sample - loss: 3.8069\n", + "9908/9908 [==============================] - 0s 27us/sample - loss: 4.8929\n", "Epoch 20/20\n", - "988129/988129 [==============================] - 40s 41us/sample - loss: 3.7964\n", - "WARNING:tensorflow:From /Users/swc/anaconda3/lib/python3.6/site-packages/tensorflow/python/keras/engine/training_v1.py:2070: Model.state_updates (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n", - "Instructions for updating:\n", - "This property should not be used in TensorFlow 2.0, as updates are applied automatically.\n", - "(6040, 32)\n", - "(3706, 32)\n" + "9908/9908 [==============================] - 0s 27us/sample - loss: 4.7612\n", + "(46, 32)\n", + "(2175, 32)\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/yulongtsai/opt/miniconda3/envs/deep_match_38/lib/python3.8/site-packages/tensorflow/python/keras/engine/training.py:2464: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", + " warnings.warn('`Model.state_updates` will be removed in a future version. '\n" ] } ], "source": [ - "#data = pd.read_csvdata = pd.read_csv(\"./movielens_sample.txt\")\n", - "sparse_features = [\"movie_id\", \"user_id\",\n", - " \"gender\", \"age\", \"occupation\", \"zip\", \"genres\"]\n", - "SEQ_LEN = 50\n", - "negsample = 0\n", - "\n", - "# 1.Label Encoding for sparse features,and process sequence features with `gen_date_set` and `gen_model_input`\n", - "\n", - "feature_max_idx = {}\n", - "for feature in sparse_features:\n", - " lbe = LabelEncoder()\n", - " data[feature] = lbe.fit_transform(data[feature]) + 1\n", - " feature_max_idx[feature] = data[feature].max() + 1\n", - "\n", - "user_profile = data[[\"user_id\", \"gender\", \"age\", \"occupation\", \"zip\"]].drop_duplicates('user_id')\n", "\n", - "item_profile = data[[\"movie_id\"]].drop_duplicates('movie_id')\n", - "\n", - "user_profile.set_index(\"user_id\", inplace=True)\n", - "\n", - "user_item_list = data.groupby(\"user_id\")['movie_id'].apply(list)\n", - "\n", - "train_set, test_set = gen_data_set(data, SEQ_LEN, negsample)\n", - "\n", - "train_model_input, train_label = gen_model_input(train_set, user_profile, SEQ_LEN)\n", - "test_model_input, test_label = gen_model_input(test_set, user_profile, SEQ_LEN)\n", "\n", "# 2.count #unique features for each sparse field and generate feature config for sequence feature\n", "\n", @@ -239,6 +784,7 @@ " embedding_name=\"genres\"), SEQ_LEN, 'mean', 'hist_len'),\n", " ]\n", "\n", + "# item 只有 id，沒有對應的 item features\n", "item_feature_columns = [SparseFeat('movie_id', feature_max_idx['movie_id'], embedding_dim)]\n", "\n", "from collections import Counter\n", @@ -257,6 +803,7 @@ "model = YoutubeDNN(user_feature_columns, item_feature_columns, user_dnn_hidden_units=(128,64, embedding_dim), sampler_config=sampler_config)\n", "#model = MIND(user_feature_columns,item_feature_columns,dynamic_k=False,k_max=2, user_dnn_hidden_units=(128,64, embedding_dim), sampler_config=sampler_config)\n", "\n", + "# sampledsoftmaxloss - important loss function apply negative sampling for multi-classes\n", "model.compile(optimizer=\"adam\", loss=sampledsoftmaxloss)\n", "\n", "history = model.fit(train_model_input, train_label, # train_label,\n", @@ -277,6 +824,70 @@ "print(item_embs.shape)" ] }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(array([-0.06659417, 0.06063242, -0.03848748, -0.28001413, 0.03008841,\n", + " -0.02134304, -0.02421067, 0.09181335, 0.00747757, -0.03091125,\n", + " -0.07595862, -0.02044039, 0.08409872, 0.06073863, -0.1624648 ,\n", + " -0.34542832, 0.49704835, 0.23406592, -0.03664735, 0.05653051,\n", + " 0.08093211, 0.08394654, 0.17315595, -0.0608228 , 0.2691896 ,\n", + " 0.26729694, 0.11697683, 0.3412756 , 0.29095942, -0.00721809,\n", + " 0.04854968, 0.15216674], dtype=float32),\n", + " array([-0.17227949, 0.12590104, -0.11971793, -0.28937042, 0.10706939,\n", + " -0.18632875, -0.11351259, -0.01100992, 0.06729872, 0.02495824,\n", + " 0.00294563, 0.04439833, 0.07902899, 0.25324842, 0.12017913,\n", + " -0.15498243, 0.53053665, 0.13115574, -0.09373093, 0.02526599,\n", + " 0.08486939, 0.2228344 , 0.16989477, 0.08697134, 0.16997184,\n", + " 0.27082318, 0.01734922, 0.29934275, 0.19565178, 0.1241786 ,\n", + " -0.01920863, 0.18342562], dtype=float32))" + ] + }, + "execution_count": 21, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "user_embs[0], item_embs[0]" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "tf.keras.utils.plot_model(\n", + " model,\n", + " to_file=\"model.png\",\n", + " show_shapes=True,\n", + " show_dtype=False,\n", + " show_layer_names=True,\n", + " rankdir=\"TB\",\n", + " expand_nested=True,\n", + " dpi=96,layer_range=None,\n", + " # show_layer_activatinotallow=True\n", + " )" + ] + }, { "cell_type": "markdown", "metadata": { @@ -295,7 +906,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 10, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -303,23 +914,34 @@ "id": "j2ZNYNBOOqrN", "outputId": "2eec5e82-2d2b-4fe0-9b83-2a74a4dc52ba" }, + "outputs": [], + "source": [ + "# ! pip install faiss-cpu" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, "outputs": [ { - "name": "stdout", - "output_type": "stream", - "text": [ - "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n", - "Requirement already satisfied: faiss-cpu in /usr/local/lib/python3.7/dist-packages (1.7.2)\n" - ] + "data": { + "text/plain": [ + "46" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" } ], "source": [ - "! pip install faiss-cpu" + "len(test_set)" ] }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 12, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -332,7 +954,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "6040it [00:02, 2769.01it/s]" + "46it [00:00, 2016.05it/s]" ] }, { @@ -340,8 +962,8 @@ "output_type": "stream", "text": [ "\n", - "recall 0.33708609271523177\n", - "hit rate 0.33708609271523177\n" + "recall 0.15217391304347827\n", + "hit rate 0.15217391304347827\n" ] }, { @@ -364,6 +986,8 @@ "# faiss.normalize_L2(item_embs)\n", "index.add(item_embs)\n", "# faiss.normalize_L2(user_embs)\n", + "# query = user embedding\n", + "# indexes = item embedding\n", "D, I = index.search(np.ascontiguousarray(user_embs), 50)\n", "s = []\n", "hit = 0\n", @@ -381,6 +1005,69 @@ "print(\"recall\", np.mean(s))\n", "print(\"hit rate\", hit / len(test_user_model_input['user_id']))" ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "dict_keys(['user_id', 'movie_id', 'hist_movie_id', 'hist_genres', 'hist_len', 'genres', 'gender', 'age', 'occupation', 'zip'])" + ] + }, + "execution_count": 13, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "test_user_model_input.keys()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Q&A\n", + "\n", + "# Instance (y)\n", + "# Retrieval - which video will be watched next?\n", + "# seqeuntial features like watch histroy, search history (use mean pooling, we could use attention now)\n", + "# user-viewed (i1, i2, i3), we could produce 2 examples\n", + " # user-i1 : i2,\n", + " # user-i1,i2, : i3\n", + "\n", + "# Feature Engineering (x)\n", + "# Retrieval\n", + "# ExampleAge\n", + "# not used here, basically viewed_since_upload_hours\n", + "# inference - set to zero - user like new video, so there is a new video bias (like positional bias)\n", + "\n", + "# Ranking\n", + "# y - expected watch time per imprerssion of user-video\n", + "# time since last eatch (channel)\n", + "# num_previous impression\n", + "\n", + "# Network Arct\n", + "# Could it expand item-side information and user-side information?\n", + "# Negative Sampler - Is it actually work as loss function level\n", + " # Yes, its a loss func\n", + " # we can define the negtive sampler to give Q(y|x)\n", + "\n", + "# Serving\n", + "# Serving format of item embedding - N dimension vectors\n", + "# The Serving Format of user embedding - N dimension vectors\n", + "\n", + "# Fresh features could be streaming by realtime-data-pipeline\n", + "# Item features is more likely to be static\n", + "# New Video need a cold-start- method (find most K similar and take mean of the video)\n", + "\n", + "# Item Profile still needs to be added..." + ] } ], "metadata": { @@ -392,7 +1079,7 @@ }, "gpuClass": "standard", "kernelspec": { - "display_name": "Python 3", + "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, @@ -406,9 +1093,9 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.6.5" + "version": "3.8.19" } }, "nbformat": 4, - "nbformat_minor": 1 + "nbformat_minor": 4 } diff --git a/examples/model.png b/examples/model.png new file mode 100644 index 0000000..daae01e Binary files /dev/null and b/examples/model.png differ diff --git a/requirements.txt b/requirements.txt new file mode 100644 index 0000000..62b8fa6 --- /dev/null +++ b/requirements.txt @@ -0,0 +1,9 @@ +faiss-cpu # only for python3.8 +recommonmark==0.7.1 +tensorflow==2.6.2 +pandas==1.1.5 +numpy==1.19 # typeDict +tqdm +scikit-learn +jupyter # for EDA +-e . # deepmatch local install \ No newline at end of file