From 6331cc397931c3d17d3f6cf56f26abba4a92aa17 Mon Sep 17 00:00:00 2001 From: Hyo-kyun Park Date: Thu, 7 Mar 2019 18:38:22 +0900 Subject: [PATCH] modifications to use latest versions of python3 & pytorch --- demo1_prepare_data.ipynb | 282 +++++++++++------------- demo2_building_blocks.ipynb | 326 ++++++++++++---------------- demo3_using_it.ipynb | 417 +++++++++++++++++++++--------------- wide_deep/data_utils.py | 10 +- wide_deep/torch_model.py | 27 ++- 5 files changed, 527 insertions(+), 535 deletions(-) diff --git a/demo1_prepare_data.ipynb b/demo1_prepare_data.ipynb index 9fface4..30462f8 100644 --- a/demo1_prepare_data.ipynb +++ b/demo1_prepare_data.ipynb @@ -13,27 +13,25 @@ }, { "cell_type": "code", - "execution_count": 1, - "metadata": { - "collapsed": false - }, + "execution_count": 2, + "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", - "\n", "\n", " \n", @@ -187,7 +185,7 @@ "4 0 " ] }, - "execution_count": 1, + "execution_count": 2, "metadata": {}, "output_type": "execute_result" } @@ -220,10 +218,8 @@ }, { "cell_type": "code", - "execution_count": 2, - "metadata": { - "collapsed": true - }, + "execution_count": 3, + "metadata": {}, "outputs": [], "source": [ "wide_cols = ['age','hours_per_week','education', 'relationship','workclass',\n", @@ -247,10 +243,8 @@ }, { "cell_type": "code", - "execution_count": 3, - "metadata": { - "collapsed": true - }, + "execution_count": 4, + "metadata": {}, "outputs": [], "source": [ "# If embeddings_cols does not include the embeddings dimensions it will be set as\n", @@ -276,10 +270,8 @@ }, { "cell_type": "code", - "execution_count": 4, - "metadata": { - "collapsed": false - }, + "execution_count": 5, + "metadata": {}, "outputs": [], "source": [ "Y = np.array(DF[target])\n", @@ -306,10 +298,8 @@ }, { "cell_type": "code", - "execution_count": 5, - "metadata": { - "collapsed": false - }, + "execution_count": 6, + "metadata": {}, "outputs": [ { "data": { @@ -322,7 +312,7 @@ "Name: education_occupation, dtype: object" ] }, - "execution_count": 5, + "execution_count": 6, "metadata": {}, "output_type": "execute_result" } @@ -349,10 +339,8 @@ }, { "cell_type": "code", - "execution_count": 6, - "metadata": { - "collapsed": true - }, + "execution_count": 7, + "metadata": {}, "outputs": [], "source": [ "def label_encode(df, cols=None):\n", @@ -379,10 +367,10 @@ " val_types[c] = df[c].unique()\n", "\n", " val_to_idx = dict()\n", - " for k, v in val_types.iteritems():\n", + " for k, v in val_types.items():\n", " val_to_idx[k] = {o: i for i, o in enumerate(val_types[k])}\n", "\n", - " for k, v in val_to_idx.iteritems():\n", + " for k, v in val_to_idx.items():\n", " df[k] = df[k].apply(lambda x: v[x])\n", "\n", " return val_to_idx, df\n", @@ -392,7 +380,7 @@ "encoding_dict,df_tmp = label_encode(df_tmp)\n", "encoding_dict = {k:encoding_dict[k] for k in encoding_dict if k in deep_cols}\n", "embeddings_input = []\n", - "for k,v in encoding_dict.iteritems():\n", + "for k,v in encoding_dict.items():\n", " embeddings_input.append((k, len(v), emb_dim[k]))" ] }, @@ -405,29 +393,13 @@ }, { "cell_type": "code", - "execution_count": 7, - "metadata": { - "collapsed": false - }, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/usr/local/lib/python2.7/site-packages/sklearn/utils/validation.py:444: DataConversionWarning: Data with input dtype int64 was converted to float64 by StandardScaler.\n", - " warnings.warn(msg, DataConversionWarning)\n", - "/usr/local/lib/python2.7/site-packages/ipykernel/__main__.py:11: SettingWithCopyWarning: \n", - "A value is trying to be set on a copy of a slice from a DataFrame.\n", - "Try using .loc[row_indexer,col_indexer] = value instead\n", - "\n", - "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n" - ] - } - ], + "execution_count": 8, + "metadata": {}, + "outputs": [], "source": [ "# select the deep_cols and get the column index that will be use later\n", "# to slice the tensors\n", - "df_deep = df_tmp[deep_cols]\n", + "df_deep = df_tmp[deep_cols].copy()\n", "deep_column_idx = {k:v for v,k in enumerate(df_deep.columns)}\n", "\n", "# The continous columns will be concatenated with the embeddings, so you\n", @@ -435,13 +407,11 @@ "from sklearn.preprocessing import StandardScaler\n", "scaler = StandardScaler()\n", "for cc in continuous_cols:\n", - " df_deep[cc] = scaler.fit_transform(df_deep[cc].values.reshape(-1,1))\n", - "\n", - "df_wide = df_tmp[wide_cols+crossed_columns]\n", + " df_deep[cc] = scaler.fit_transform(df_deep[cc].values.reshape(-1,1).astype(float))\n", + "df_wide = df_tmp[wide_cols+crossed_columns]#.copy()\n", "del(df_tmp)\n", - "\n", "dummy_cols = [c for c in wide_cols+crossed_columns if c in categorical_columns]\n", - "df_wide = pd.get_dummies(df_wide, columns=dummy_cols)" + "df_wide = pd.get_dummies(df_wide, columns=dummy_cols)#.copy()\n" ] }, { @@ -455,10 +425,8 @@ }, { "cell_type": "code", - "execution_count": 8, - "metadata": { - "collapsed": false - }, + "execution_count": 9, + "metadata": {}, "outputs": [], "source": [ "from sklearn.model_selection import train_test_split\n", @@ -482,10 +450,8 @@ }, { "cell_type": "code", - "execution_count": 9, - "metadata": { - "collapsed": false - }, + "execution_count": 10, + "metadata": {}, "outputs": [ { "name": "stdout", @@ -494,7 +460,7 @@ "train_dataset(wide=array([[46, 50, 0, ..., 0, 0, 0],\n", " [32, 45, 1, ..., 0, 0, 0],\n", " [30, 30, 0, ..., 0, 0, 0],\n", - " ..., \n", + " ...,\n", " [40, 40, 0, ..., 0, 0, 0],\n", " [45, 37, 1, ..., 0, 0, 0],\n", " [40, 45, 1, ..., 0, 0, 0]]), deep=array([[ 3. , 1. , 6. , ..., 0. ,\n", @@ -503,7 +469,7 @@ " -0.48456647, 0.36942139],\n", " [ 1. , 4. , 2. , ..., 0. ,\n", " -0.63044146, -0.84110367],\n", - " ..., \n", + " ...,\n", " [ 1. , 0. , 2. , ..., 0. ,\n", " 0.09893348, -0.03408696],\n", " [ 0. , 1. , 2. , ..., 0. ,\n", @@ -519,16 +485,14 @@ }, { "cell_type": "code", - "execution_count": 10, - "metadata": { - "collapsed": false - }, + "execution_count": 11, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "[('workclass', 9, 10), ('education', 16, 10), ('native_country', 42, 12), ('relationship', 6, 8), ('occupation', 15, 10)]\n" + "[('education', 16, 10), ('relationship', 6, 8), ('native_country', 42, 12), ('workclass', 9, 10), ('occupation', 15, 10)]\n" ] } ], @@ -538,16 +502,14 @@ }, { "cell_type": "code", - "execution_count": 11, - "metadata": { - "collapsed": false - }, + "execution_count": 12, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "{'hours_per_week': 6, 'native_country': 4, 'relationship': 1, 'age': 5, 'workclass': 2, 'education': 0, 'occupation': 3}\n" + "{'education': 0, 'relationship': 1, 'workclass': 2, 'occupation': 3, 'native_country': 4, 'age': 5, 'hours_per_week': 6}\n" ] } ], @@ -557,105 +519,103 @@ }, { "cell_type": "code", - "execution_count": 14, - "metadata": { - "collapsed": false - }, + "execution_count": 13, + "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "{'education': {'10th': 12,\n", + "{'education': {'Bachelors': 0,\n", + " 'HS-grad': 1,\n", " '11th': 2,\n", - " '12th': 15,\n", - " '1st-4th': 13,\n", - " '5th-6th': 11,\n", - " '7th-8th': 8,\n", + " 'Masters': 3,\n", " '9th': 4,\n", + " 'Some-college': 5,\n", " 'Assoc-acdm': 6,\n", " 'Assoc-voc': 7,\n", - " 'Bachelors': 0,\n", + " '7th-8th': 8,\n", " 'Doctorate': 9,\n", - " 'HS-grad': 1,\n", - " 'Masters': 3,\n", - " 'Preschool': 14,\n", " 'Prof-school': 10,\n", - " 'Some-college': 5},\n", - " 'native_country': {'?': 4,\n", - " 'Cambodia': 17,\n", + " '5th-6th': 11,\n", + " '10th': 12,\n", + " '1st-4th': 13,\n", + " 'Preschool': 14,\n", + " '12th': 15},\n", + " 'relationship': {'Not-in-family': 0,\n", + " 'Husband': 1,\n", + " 'Wife': 2,\n", + " 'Own-child': 3,\n", + " 'Unmarried': 4,\n", + " 'Other-relative': 5},\n", + " 'native_country': {'United-States': 0,\n", + " 'Cuba': 1,\n", + " 'Jamaica': 2,\n", + " 'India': 3,\n", + " '?': 4,\n", + " 'Mexico': 5,\n", + " 'South': 6,\n", + " 'Puerto-Rico': 7,\n", + " 'Honduras': 8,\n", + " 'England': 9,\n", " 'Canada': 10,\n", - " 'China': 28,\n", + " 'Germany': 11,\n", + " 'Iran': 12,\n", + " 'Philippines': 13,\n", + " 'Italy': 14,\n", + " 'Poland': 15,\n", " 'Columbia': 16,\n", - " 'Cuba': 1,\n", - " 'Dominican-Republic': 24,\n", + " 'Cambodia': 17,\n", + " 'Thailand': 18,\n", " 'Ecuador': 19,\n", + " 'Laos': 20,\n", + " 'Taiwan': 21,\n", + " 'Haiti': 22,\n", + " 'Portugal': 23,\n", + " 'Dominican-Republic': 24,\n", " 'El-Salvador': 25,\n", - " 'England': 9,\n", " 'France': 26,\n", - " 'Germany': 11,\n", - " 'Greece': 35,\n", " 'Guatemala': 27,\n", - " 'Haiti': 22,\n", - " 'Holand-Netherlands': 41,\n", - " 'Honduras': 8,\n", - " 'Hong': 38,\n", - " 'Hungary': 40,\n", - " 'India': 3,\n", - " 'Iran': 12,\n", - " 'Ireland': 39,\n", - " 'Italy': 14,\n", - " 'Jamaica': 2,\n", + " 'China': 28,\n", " 'Japan': 29,\n", - " 'Laos': 20,\n", - " 'Mexico': 5,\n", - " 'Nicaragua': 36,\n", - " 'Outlying-US(Guam-USVI-etc)': 32,\n", + " 'Yugoslavia': 30,\n", " 'Peru': 31,\n", - " 'Philippines': 13,\n", - " 'Poland': 15,\n", - " 'Portugal': 23,\n", - " 'Puerto-Rico': 7,\n", + " 'Outlying-US(Guam-USVI-etc)': 32,\n", " 'Scotland': 33,\n", - " 'South': 6,\n", - " 'Taiwan': 21,\n", - " 'Thailand': 18,\n", " 'Trinadad&Tobago': 34,\n", - " 'United-States': 0,\n", + " 'Greece': 35,\n", + " 'Nicaragua': 36,\n", " 'Vietnam': 37,\n", - " 'Yugoslavia': 30},\n", - " 'occupation': {'?': 11,\n", - " 'Adm-clerical': 0,\n", - " 'Armed-Forces': 13,\n", - " 'Craft-repair': 6,\n", + " 'Hong': 38,\n", + " 'Ireland': 39,\n", + " 'Hungary': 40,\n", + " 'Holand-Netherlands': 41},\n", + " 'workclass': {'State-gov': 0,\n", + " 'Self-emp-not-inc': 1,\n", + " 'Private': 2,\n", + " 'Federal-gov': 3,\n", + " 'Local-gov': 4,\n", + " '?': 5,\n", + " 'Self-emp-inc': 6,\n", + " 'Without-pay': 7,\n", + " 'Never-worked': 8},\n", + " 'occupation': {'Adm-clerical': 0,\n", " 'Exec-managerial': 1,\n", - " 'Farming-fishing': 8,\n", " 'Handlers-cleaners': 2,\n", - " 'Machine-op-inspct': 9,\n", - " 'Other-service': 4,\n", - " 'Priv-house-serv': 14,\n", " 'Prof-specialty': 3,\n", - " 'Protective-serv': 12,\n", + " 'Other-service': 4,\n", " 'Sales': 5,\n", + " 'Craft-repair': 6,\n", + " 'Transport-moving': 7,\n", + " 'Farming-fishing': 8,\n", + " 'Machine-op-inspct': 9,\n", " 'Tech-support': 10,\n", - " 'Transport-moving': 7},\n", - " 'relationship': {'Husband': 1,\n", - " 'Not-in-family': 0,\n", - " 'Other-relative': 5,\n", - " 'Own-child': 3,\n", - " 'Unmarried': 4,\n", - " 'Wife': 2},\n", - " 'workclass': {'?': 5,\n", - " 'Federal-gov': 3,\n", - " 'Local-gov': 4,\n", - " 'Never-worked': 8,\n", - " 'Private': 2,\n", - " 'Self-emp-inc': 6,\n", - " 'Self-emp-not-inc': 1,\n", - " 'State-gov': 0,\n", - " 'Without-pay': 7}}" + " '?': 11,\n", + " 'Protective-serv': 12,\n", + " 'Armed-Forces': 13,\n", + " 'Priv-house-serv': 14}}" ] }, - "execution_count": 14, + "execution_count": 13, "metadata": {}, "output_type": "execute_result" } @@ -673,10 +633,8 @@ }, { "cell_type": "code", - "execution_count": 15, - "metadata": { - "collapsed": true - }, + "execution_count": 14, + "metadata": {}, "outputs": [], "source": [ "from wide_deep.data_utils import prepare_data" @@ -692,23 +650,23 @@ ], "metadata": { "kernelspec": { - "display_name": "Python 2", + "display_name": "Python 3", "language": "python", - "name": "python2" + "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", - "version": 2 + "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", - "pygments_lexer": "ipython2", - "version": "2.7.13" + "pygments_lexer": "ipython3", + "version": "3.6.5" } }, "nbformat": 4, - "nbformat_minor": 0 + "nbformat_minor": 1 } diff --git a/demo2_building_blocks.ipynb b/demo2_building_blocks.ipynb index afed301..6344eec 100644 --- a/demo2_building_blocks.ipynb +++ b/demo2_building_blocks.ipynb @@ -20,10 +20,8 @@ }, { "cell_type": "code", - "execution_count": 26, - "metadata": { - "collapsed": false - }, + "execution_count": 1, + "metadata": {}, "outputs": [], "source": [ "from __future__ import print_function\n", @@ -55,10 +53,8 @@ }, { "cell_type": "code", - "execution_count": 27, - "metadata": { - "collapsed": false - }, + "execution_count": 2, + "metadata": {}, "outputs": [ { "data": { @@ -66,7 +62,7 @@ "train_dataset(wide=array([[46, 50, 0, ..., 0, 0, 0],\n", " [32, 45, 1, ..., 0, 0, 0],\n", " [30, 30, 0, ..., 0, 0, 0],\n", - " ..., \n", + " ...,\n", " [40, 40, 0, ..., 0, 0, 0],\n", " [45, 37, 1, ..., 0, 0, 0],\n", " [40, 45, 1, ..., 0, 0, 0]]), deep=array([[ 3. , 1. , 6. , ..., 0. ,\n", @@ -75,7 +71,7 @@ " -0.48456647, 0.36942139],\n", " [ 1. , 4. , 2. , ..., 0. ,\n", " -0.63044146, -0.84110367],\n", - " ..., \n", + " ...,\n", " [ 1. , 0. , 2. , ..., 0. ,\n", " 0.09893348, -0.03408696],\n", " [ 0. , 1. , 2. , ..., 0. ,\n", @@ -84,7 +80,7 @@ " 0.09893348, 0.36942139]]), labels=array([1, 0, 0, ..., 0, 0, 0]))" ] }, - "execution_count": 27, + "execution_count": 2, "metadata": {}, "output_type": "execute_result" } @@ -104,32 +100,13 @@ }, { "cell_type": "code", - "execution_count": 28, - "metadata": { - "collapsed": false - }, - "outputs": [ - { - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAB3gAAAGQCAYAAABMP7EKAAAME2lDQ1BJQ0MgUHJvZmlsZQAASImV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\n", - "text/plain": [ - "" - ] - }, - "execution_count": 28, - "metadata": { - "image/png": { - "height": 500, - "width": 1000 - } - }, - "output_type": "execute_result" - } - ], + "execution_count": 3, + "metadata": {}, + "outputs": [], "source": [ - "from IPython.display import Image\n", - "PATH = \"/Users/javier/Desktop/wide_deep.png\"\n", - "Image(filename = PATH, width=1000, height=500)" + "# from IPython.display import Image\n", + "# PATH = \"/Users/javier/Desktop/wide_deep.png\"\n", + "# Image(filename = PATH, width=1000, height=500)" ] }, { @@ -160,16 +137,14 @@ }, { "cell_type": "code", - "execution_count": 29, - "metadata": { - "collapsed": false - }, + "execution_count": 4, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Linear (798 -> 1)\n" + "Linear(in_features=798, out_features=1, bias=True)\n" ] } ], @@ -193,10 +168,8 @@ }, { "cell_type": "code", - "execution_count": 30, - "metadata": { - "collapsed": true - }, + "execution_count": 5, + "metadata": {}, "outputs": [], "source": [ "import torch\n", @@ -227,17 +200,15 @@ "\n", " def forward(self,X):\n", "\n", - " out = F.sigmoid(self.linear(X))\n", + " out = torch.sigmoid(self.linear(X))\n", "\n", " return out\n" ] }, { "cell_type": "code", - "execution_count": 31, - "metadata": { - "collapsed": false - }, + "execution_count": 6, + "metadata": {}, "outputs": [], "source": [ "wide_dim = wd_dataset['train_dataset'].wide.shape[1]\n", @@ -247,17 +218,15 @@ }, { "cell_type": "code", - "execution_count": 32, - "metadata": { - "collapsed": false - }, + "execution_count": 7, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Wide (\n", - " (linear): Linear (798 -> 1)\n", + "Wide(\n", + " (linear): Linear(in_features=798, out_features=1, bias=True)\n", ")\n" ] } @@ -277,10 +246,8 @@ }, { "cell_type": "code", - "execution_count": 33, - "metadata": { - "collapsed": false - }, + "execution_count": 8, + "metadata": {}, "outputs": [ { "data": { @@ -288,7 +255,7 @@ "(34189, 1)" ] }, - "execution_count": 33, + "execution_count": 8, "metadata": {}, "output_type": "execute_result" } @@ -299,10 +266,8 @@ }, { "cell_type": "code", - "execution_count": 34, - "metadata": { - "collapsed": false - }, + "execution_count": 9, + "metadata": {}, "outputs": [ { "data": { @@ -310,13 +275,13 @@ "array([[ 1, 46, 50, ..., 0, 0, 0],\n", " [ 0, 32, 45, ..., 0, 0, 0],\n", " [ 0, 30, 30, ..., 0, 0, 0],\n", - " ..., \n", + " ...,\n", " [ 0, 40, 40, ..., 0, 0, 0],\n", " [ 0, 45, 37, ..., 0, 0, 0],\n", " [ 0, 40, 45, ..., 0, 0, 0]])" ] }, - "execution_count": 34, + "execution_count": 9, "metadata": {}, "output_type": "execute_result" } @@ -337,25 +302,33 @@ }, { "cell_type": "code", - "execution_count": 35, - "metadata": { - "collapsed": false - }, + "execution_count": 10, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Epoch 1 of 10, Loss: 0.378, accuracy: 0.7799\n", - "Epoch 2 of 10, Loss: 0.622, accuracy: 0.819\n", - "Epoch 3 of 10, Loss: 0.483, accuracy: 0.8264\n", - "Epoch 4 of 10, Loss: 0.46, accuracy: 0.83\n", - "Epoch 5 of 10, Loss: 0.378, accuracy: 0.8329\n", - "Epoch 6 of 10, Loss: 0.126, accuracy: 0.8337\n", - "Epoch 7 of 10, Loss: 0.318, accuracy: 0.8354\n", - "Epoch 8 of 10, Loss: 0.311, accuracy: 0.8346\n", - "Epoch 9 of 10, Loss: 0.262, accuracy: 0.8354\n", - "Epoch 10 of 10, Loss: 0.38, accuracy: 0.8361\n" + "26682.0 34189\n", + "Epoch 1 of 10, Loss: 0.227, accuracy: 0.7804\n", + "28058.0 34189\n", + "Epoch 2 of 10, Loss: 0.374, accuracy: 0.8207\n", + "28261.0 34189\n", + "Epoch 3 of 10, Loss: 0.473, accuracy: 0.8266\n", + "28373.0 34189\n", + "Epoch 4 of 10, Loss: 0.17, accuracy: 0.8299\n", + "28475.0 34189\n", + "Epoch 5 of 10, Loss: 0.344, accuracy: 0.8329\n", + "28499.0 34189\n", + "Epoch 6 of 10, Loss: 0.573, accuracy: 0.8336\n", + "28530.0 34189\n", + "Epoch 7 of 10, Loss: 0.328, accuracy: 0.8345\n", + "28544.0 34189\n", + "Epoch 8 of 10, Loss: 0.301, accuracy: 0.8349\n", + "28568.0 34189\n", + "Epoch 9 of 10, Loss: 0.21, accuracy: 0.8356\n", + "28587.0 34189\n", + "Epoch 10 of 10, Loss: 0.248, accuracy: 0.8361\n" ] } ], @@ -377,16 +350,17 @@ "\n", " optimizer.zero_grad()\n", " y_pred = wide_model(X_w)\n", - " loss = F.binary_cross_entropy(y_pred, y)\n", + " loss = F.binary_cross_entropy(y_pred, y.view(-1, 1))\n", " loss.backward()\n", " optimizer.step()\n", "\n", " total+= y.size(0)\n", " y_pred_cat = (y_pred > 0.5).squeeze(1).float()\n", - " correct+= float((y_pred_cat == y).sum().data[0])\n", - "\n", + " \n", + " correct+= float((y_pred_cat == y).sum().item())\n", + " print (correct, total)\n", " print ('Epoch {} of {}, Loss: {}, accuracy: {}'.format(epoch+1,\n", - " n_epochs, round(loss.data[0],3), round(correct/total,4)))" + " n_epochs, round(loss.item(),3), round(correct/total,4)))" ] }, { @@ -404,17 +378,15 @@ }, { "cell_type": "code", - "execution_count": 36, - "metadata": { - "collapsed": false - }, + "execution_count": 11, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "[('workclass', 9, 10), ('education', 16, 10), ('native_country', 42, 10), ('relationship', 6, 8), ('occupation', 15, 10)]\n", - "{'hours_per_week': 6, 'native_country': 4, 'relationship': 1, 'age': 5, 'workclass': 2, 'education': 0, 'occupation': 3}\n" + "[('education', 16, 10), ('occupation', 15, 10), ('relationship', 6, 8), ('native_country', 42, 10), ('workclass', 9, 10)]\n", + "{'education': 0, 'relationship': 1, 'workclass': 2, 'occupation': 3, 'native_country': 4, 'age': 5, 'hours_per_week': 6}\n" ] } ], @@ -434,16 +406,14 @@ }, { "cell_type": "code", - "execution_count": 37, - "metadata": { - "collapsed": false - }, + "execution_count": 12, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Embedding(9, 10)\n" + "Embedding(16, 10)\n" ] } ], @@ -462,10 +432,8 @@ }, { "cell_type": "code", - "execution_count": 38, - "metadata": { - "collapsed": false - }, + "execution_count": 13, + "metadata": {}, "outputs": [], "source": [ "class Deep(nn.Module):\n", @@ -519,7 +487,7 @@ " for i in range(1,len(self.hidden_layers)):\n", " x_deep = F.relu( getattr(self, 'linear_'+str(i+1))(x_deep) )\n", "\n", - " out = F.sigmoid(self.output(x_deep))\n", + " out = torch.sigmoid(self.output(x_deep))\n", "\n", " return out\n" ] @@ -533,10 +501,8 @@ }, { "cell_type": "code", - "execution_count": 39, - "metadata": { - "collapsed": false - }, + "execution_count": 14, + "metadata": {}, "outputs": [], "source": [ "deep_column_idx = wd_dataset['deep_column_idx']\n", @@ -547,24 +513,22 @@ }, { "cell_type": "code", - "execution_count": 40, - "metadata": { - "collapsed": false - }, + "execution_count": 15, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Deep (\n", - " (emb_layer_workclass): Embedding(9, 10)\n", + "Deep(\n", " (emb_layer_education): Embedding(16, 10)\n", - " (emb_layer_native_country): Embedding(42, 10)\n", - " (emb_layer_relationship): Embedding(6, 8)\n", " (emb_layer_occupation): Embedding(15, 10)\n", - " (linear_1): Linear (50 -> 100)\n", - " (linear_2): Linear (100 -> 50)\n", - " (output): Linear (50 -> 1)\n", + " (emb_layer_relationship): Embedding(6, 8)\n", + " (emb_layer_native_country): Embedding(42, 10)\n", + " (emb_layer_workclass): Embedding(9, 10)\n", + " (linear_1): Linear(in_features=50, out_features=100, bias=True)\n", + " (linear_2): Linear(in_features=100, out_features=50, bias=True)\n", + " (output): Linear(in_features=50, out_features=1, bias=True)\n", ")\n" ] } @@ -589,10 +553,8 @@ }, { "cell_type": "code", - "execution_count": 41, - "metadata": { - "collapsed": false - }, + "execution_count": 16, + "metadata": {}, "outputs": [ { "data": { @@ -603,7 +565,7 @@ " -0.48456647, 0.36942139],\n", " [ 0. , 1. , 4. , ..., 0. ,\n", " -0.63044146, -0.84110367],\n", - " ..., \n", + " ...,\n", " [ 0. , 1. , 0. , ..., 0. ,\n", " 0.09893348, -0.03408696],\n", " [ 0. , 0. , 1. , ..., 0. ,\n", @@ -612,7 +574,7 @@ " 0.09893348, 0.36942139]])" ] }, - "execution_count": 41, + "execution_count": 16, "metadata": {}, "output_type": "execute_result" } @@ -624,25 +586,23 @@ }, { "cell_type": "code", - "execution_count": 42, - "metadata": { - "collapsed": false - }, + "execution_count": 17, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Epoch 1 of 10, Loss: 0.314, accuracy: 0.8225\n", - "Epoch 2 of 10, Loss: 0.444, accuracy: 0.8373\n", - "Epoch 3 of 10, Loss: 0.448, accuracy: 0.8417\n", - "Epoch 4 of 10, Loss: 0.264, accuracy: 0.8418\n", - "Epoch 5 of 10, Loss: 0.146, accuracy: 0.8443\n", - "Epoch 6 of 10, Loss: 0.4, accuracy: 0.8449\n", - "Epoch 7 of 10, Loss: 0.52, accuracy: 0.8459\n", - "Epoch 8 of 10, Loss: 0.428, accuracy: 0.8472\n", - "Epoch 9 of 10, Loss: 0.565, accuracy: 0.8453\n", - "Epoch 10 of 10, Loss: 0.27, accuracy: 0.8474\n" + "Epoch 1 of 10, Loss: 0.433, accuracy: 0.8252\n", + "Epoch 2 of 10, Loss: 0.487, accuracy: 0.8395\n", + "Epoch 3 of 10, Loss: 0.235, accuracy: 0.8427\n", + "Epoch 4 of 10, Loss: 0.706, accuracy: 0.8426\n", + "Epoch 5 of 10, Loss: 0.425, accuracy: 0.8443\n", + "Epoch 6 of 10, Loss: 0.495, accuracy: 0.8456\n", + "Epoch 7 of 10, Loss: 0.291, accuracy: 0.8463\n", + "Epoch 8 of 10, Loss: 0.181, accuracy: 0.8455\n", + "Epoch 9 of 10, Loss: 0.362, accuracy: 0.8474\n", + "Epoch 10 of 10, Loss: 0.158, accuracy: 0.8468\n" ] } ], @@ -664,16 +624,16 @@ "\n", " optimizer.zero_grad()\n", " y_pred = deep_model(X_d)\n", - " loss = F.binary_cross_entropy(y_pred, y)\n", + " loss = F.binary_cross_entropy(y_pred, y.view(-1, 1))\n", " loss.backward()\n", " optimizer.step()\n", "\n", " total+= y.size(0)\n", " y_pred_cat = (y_pred > 0.5).squeeze(1).float()\n", - " correct+= float((y_pred_cat == y).sum().data[0])\n", - "\n", + " correct+= float((y_pred_cat == y).sum().item())\n", + " \n", " print ('Epoch {} of {}, Loss: {}, accuracy: {}'.format(epoch+1,\n", - " n_epochs, round(loss.data[0],3), round(correct/total,4)))" + " n_epochs, round(loss.item(),3), round(correct/total,4)))" ] }, { @@ -698,10 +658,8 @@ }, { "cell_type": "code", - "execution_count": 43, - "metadata": { - "collapsed": true - }, + "execution_count": 18, + "metadata": {}, "outputs": [], "source": [ "class WideDeep(nn.Module):\n", @@ -742,7 +700,7 @@ "\n", " wide_deep_input = torch.cat([x_deep, X_w.float()], 1)\n", "\n", - " out = F.sigmoid(self.output(wide_deep_input))\n", + " out = torch.sigmoid(self.output(wide_deep_input))\n", "\n", " return out" ] @@ -756,10 +714,8 @@ }, { "cell_type": "code", - "execution_count": 44, - "metadata": { - "collapsed": false - }, + "execution_count": 19, + "metadata": {}, "outputs": [], "source": [ "wide_deep_model = WideDeep(wide_dim, embeddings_input, continuous_cols, deep_column_idx, hidden_layers, n_class)" @@ -767,27 +723,25 @@ }, { "cell_type": "code", - "execution_count": 45, - "metadata": { - "collapsed": false - }, + "execution_count": 20, + "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "WideDeep (\n", - " (emb_layer_workclass): Embedding(9, 10)\n", + "WideDeep(\n", " (emb_layer_education): Embedding(16, 10)\n", - " (emb_layer_native_country): Embedding(42, 10)\n", - " (emb_layer_relationship): Embedding(6, 8)\n", " (emb_layer_occupation): Embedding(15, 10)\n", - " (linear_1): Linear (50 -> 100)\n", - " (linear_2): Linear (100 -> 50)\n", - " (output): Linear (848 -> 1)\n", + " (emb_layer_relationship): Embedding(6, 8)\n", + " (emb_layer_native_country): Embedding(42, 10)\n", + " (emb_layer_workclass): Embedding(9, 10)\n", + " (linear_1): Linear(in_features=50, out_features=100, bias=True)\n", + " (linear_2): Linear(in_features=100, out_features=50, bias=True)\n", + " (output): Linear(in_features=848, out_features=1, bias=True)\n", ")" ] }, - "execution_count": 45, + "execution_count": 20, "metadata": {}, "output_type": "execute_result" } @@ -807,10 +761,8 @@ }, { "cell_type": "code", - "execution_count": 46, - "metadata": { - "collapsed": false - }, + "execution_count": 21, + "metadata": {}, "outputs": [], "source": [ "class WideDeepLoader(Dataset):\n", @@ -854,25 +806,23 @@ }, { "cell_type": "code", - "execution_count": 47, - "metadata": { - "collapsed": false - }, + "execution_count": 22, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Epoch 1 of 10, Loss: 0.502, accuracy: 0.824\n", - "Epoch 2 of 10, Loss: 0.472, accuracy: 0.8377\n", - "Epoch 3 of 10, Loss: 0.463, accuracy: 0.8405\n", - "Epoch 4 of 10, Loss: 0.096, accuracy: 0.8418\n", - "Epoch 5 of 10, Loss: 0.454, accuracy: 0.8426\n", - "Epoch 6 of 10, Loss: 0.235, accuracy: 0.8441\n", - "Epoch 7 of 10, Loss: 0.302, accuracy: 0.8452\n", - "Epoch 8 of 10, Loss: 0.342, accuracy: 0.845\n", - "Epoch 9 of 10, Loss: 0.405, accuracy: 0.8453\n", - "Epoch 10 of 10, Loss: 0.23, accuracy: 0.8477\n" + "Epoch 1 of 10, Loss: 0.144, accuracy: 0.8245\n", + "Epoch 2 of 10, Loss: 0.316, accuracy: 0.8384\n", + "Epoch 3 of 10, Loss: 0.262, accuracy: 0.8416\n", + "Epoch 4 of 10, Loss: 0.219, accuracy: 0.8421\n", + "Epoch 5 of 10, Loss: 0.311, accuracy: 0.8433\n", + "Epoch 6 of 10, Loss: 0.303, accuracy: 0.8439\n", + "Epoch 7 of 10, Loss: 0.478, accuracy: 0.844\n", + "Epoch 8 of 10, Loss: 0.366, accuracy: 0.8451\n", + "Epoch 9 of 10, Loss: 0.356, accuracy: 0.8477\n", + "Epoch 10 of 10, Loss: 0.544, accuracy: 0.8459\n" ] } ], @@ -891,16 +841,16 @@ "\n", " optimizer.zero_grad()\n", " y_pred = wide_deep_model(X_w, X_d)\n", - " loss = F.binary_cross_entropy(y_pred, y)\n", + " loss = F.binary_cross_entropy(y_pred, y.view(-1,1))\n", " loss.backward()\n", " optimizer.step()\n", "\n", " total+= y.size(0)\n", " y_pred_cat = (y_pred > 0.5).squeeze(1).float()\n", - " correct+= float((y_pred_cat == y).sum().data[0])\n", + " correct+= float((y_pred_cat == y).sum().item())\n", "\n", " print ('Epoch {} of {}, Loss: {}, accuracy: {}'.format(epoch+1,\n", - " n_epochs, round(loss.data[0],3), round(correct/total,4)))\n" + " n_epochs, round(loss.item(),3), round(correct/total,4)))\n" ] }, { @@ -917,23 +867,23 @@ ], "metadata": { "kernelspec": { - "display_name": "Python 2", + "display_name": "Python 3", "language": "python", - "name": "python2" + "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", - "version": 2 + "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", - "pygments_lexer": "ipython2", - "version": "2.7.13" + "pygments_lexer": "ipython3", + "version": "3.6.5" } }, "nbformat": 4, - "nbformat_minor": 0 + "nbformat_minor": 1 } diff --git a/demo3_using_it.ipynb b/demo3_using_it.ipynb index 5440400..ad04b08 100644 --- a/demo3_using_it.ipynb +++ b/demo3_using_it.ipynb @@ -22,27 +22,25 @@ }, { "cell_type": "code", - "execution_count": 8, - "metadata": { - "collapsed": false - }, + "execution_count": 1, + "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", - "\n", "
\n", " \n", @@ -196,7 +194,7 @@ "4 0 " ] }, - "execution_count": 8, + "execution_count": 1, "metadata": {}, "output_type": "execute_result" } @@ -227,10 +225,8 @@ }, { "cell_type": "code", - "execution_count": 9, - "metadata": { - "collapsed": true - }, + "execution_count": 2, + "metadata": {}, "outputs": [], "source": [ "# Let's define a target for logistic regression:\n", @@ -256,10 +252,8 @@ }, { "cell_type": "code", - "execution_count": 10, - "metadata": { - "collapsed": false - }, + "execution_count": 3, + "metadata": {}, "outputs": [], "source": [ "from wide_deep.data_utils import prepare_data\n", @@ -277,10 +271,8 @@ }, { "cell_type": "code", - "execution_count": 11, - "metadata": { - "collapsed": false - }, + "execution_count": 4, + "metadata": {}, "outputs": [], "source": [ "# Network set up\n", @@ -309,24 +301,22 @@ }, { "cell_type": "code", - "execution_count": 12, - "metadata": { - "collapsed": false - }, + "execution_count": 5, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "WideDeep (\n", - " (emb_layer_workclass): Embedding(9, 10)\n", - " (emb_layer_education): Embedding(16, 10)\n", + "WideDeep(\n", " (emb_layer_native_country): Embedding(42, 10)\n", " (emb_layer_relationship): Embedding(6, 8)\n", " (emb_layer_occupation): Embedding(15, 10)\n", - " (linear_1): Linear (50 -> 100)\n", - " (linear_2): Linear (100 -> 50)\n", - " (output): Linear (848 -> 1)\n", + " (emb_layer_education): Embedding(16, 10)\n", + " (emb_layer_workclass): Embedding(9, 10)\n", + " (linear_1): Linear(in_features=50, out_features=100, bias=True)\n", + " (linear_2): Linear(in_features=100, out_features=50, bias=True)\n", + " (output): Linear(in_features=848, out_features=1, bias=True)\n", ")\n" ] } @@ -344,36 +334,24 @@ }, { "cell_type": "code", - "execution_count": 13, - "metadata": { - "collapsed": false - }, + "execution_count": 6, + "metadata": {}, "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/usr/local/lib/python2.7/site-packages/torch/nn/functional.py:767: UserWarning: Using a target size (torch.Size([64])) that is different to the input size (torch.Size([64, 1])) is deprecated. Please ensure they have the same size.\n", - " \"Please ensure they have the same size.\".format(target.size(), input.size()))\n", - "/usr/local/lib/python2.7/site-packages/torch/nn/functional.py:767: UserWarning: Using a target size (torch.Size([13])) that is different to the input size (torch.Size([13, 1])) is deprecated. Please ensure they have the same size.\n", - " \"Please ensure they have the same size.\".format(target.size(), input.size()))\n" - ] - }, { "name": "stdout", "output_type": "stream", "text": [ - "Epoch 1 of 10, Loss: 0.215, accuracy: 0.8175\n", - "Epoch 2 of 10, Loss: 0.356, accuracy: 0.8396\n", - "Epoch 3 of 10, Loss: 0.229, accuracy: 0.842\n", - "Epoch 4 of 10, Loss: 0.531, accuracy: 0.8425\n", - "Epoch 5 of 10, Loss: 0.197, accuracy: 0.8438\n", - "Epoch 6 of 10, Loss: 0.134, accuracy: 0.844\n", - "Epoch 7 of 10, Loss: 0.454, accuracy: 0.8463\n", - "Epoch 8 of 10, Loss: 0.156, accuracy: 0.8464\n", - "Epoch 9 of 10, Loss: 0.217, accuracy: 0.8452\n", - "Epoch 10 of 10, Loss: 0.445, accuracy: 0.8472\n", - "0.838258377124\n" + "Epoch 1 of 10, Loss: 0.136, accuracy: 0.8246\n", + "Epoch 2 of 10, Loss: 0.106, accuracy: 0.8392\n", + "Epoch 3 of 10, Loss: 0.513, accuracy: 0.8421\n", + "Epoch 4 of 10, Loss: 0.345, accuracy: 0.8414\n", + "Epoch 5 of 10, Loss: 0.29, accuracy: 0.843\n", + "Epoch 6 of 10, Loss: 0.227, accuracy: 0.8443\n", + "Epoch 7 of 10, Loss: 0.426, accuracy: 0.845\n", + "Epoch 8 of 10, Loss: 0.183, accuracy: 0.8454\n", + "Epoch 9 of 10, Loss: 0.322, accuracy: 0.8461\n", + "Epoch 10 of 10, Loss: 0.246, accuracy: 0.8469\n", + "0.8382583771241384\n" ] } ], @@ -398,49 +376,63 @@ }, { "cell_type": "code", - "execution_count": 14, - "metadata": { - "collapsed": false - }, + "execution_count": 7, + "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "{'10th': array([-0.18979575, 1.4436841 , -0.50139612, -0.85227281, 1.36461151,\n", - " 0.3559041 , -0.58077377, 0.57836998, 0.09822965, 0.45356399], dtype=float32),\n", - " '11th': array([ 0.45051831, -1.17895794, -0.70969492, -0.41443011, -0.54592711,\n", - " 2.06732845, 0.97312623, -1.66578746, 0.15288909, -0.13219695], dtype=float32),\n", - " '12th': array([-0.55539042, 1.34430635, -0.14818592, -1.01501787, -1.85061646,\n", - " -1.42545903, 0.30155715, 1.02573991, -0.42215505, 1.02378154], dtype=float32),\n", - " '1st-4th': array([ 1.84887922, 1.20987594, 0.2984882 , -1.79686284, 0.59199595,\n", - " -0.09441201, -0.26749009, 0.20149775, -0.73544145, -0.51700133], dtype=float32),\n", - " '5th-6th': array([-0.05392418, 0.36236417, 0.47461176, 0.41363204, -0.2278301 ,\n", - " -0.5376063 , 2.63320708, 2.04696202, -0.49895033, -0.29155737], dtype=float32),\n", - " '7th-8th': array([ 0.12547047, 0.05075515, -1.44649279, -1.56195939, -1.32460868,\n", - " -0.34222227, 0.88958579, 0.47252822, -0.09495597, -0.02843619], dtype=float32),\n", - " '9th': array([-0.14014393, 1.28053474, 0.35706842, -1.89409554, -0.80370718,\n", - " -0.53732723, 0.39302668, 0.40100414, 0.96709979, -0.54595846], dtype=float32),\n", - " 'Assoc-acdm': array([ 2.23356485, 1.00816226, -2.27655983, 0.9915536 , -0.55686516,\n", - " 1.46899855, -1.43701446, -0.46746022, -0.05142261, 0.58451122], dtype=float32),\n", - " 'Assoc-voc': array([ 1.11377907, 0.41131377, -1.40442908, 0.05879473, -0.13471135,\n", - " -0.37147653, -0.39430454, -0.38298509, 0.09182382, -0.18972228], dtype=float32),\n", - " 'Bachelors': array([-2.05170012, -1.35262275, 1.57654059, 0.70553464, 0.80315828,\n", - " -2.61309099, 0.46207047, 0.25938991, 1.27118778, -2.02918983], dtype=float32),\n", - " 'Doctorate': array([-1.51427305, -0.57435513, 0.27152076, 3.68988252, -0.92434132,\n", - " 0.86953694, -0.62879491, -0.6649121 , 0.37984002, 0.42188498], dtype=float32),\n", - " 'HS-grad': array([ 0.5675661 , -0.16484176, -0.25774002, 0.12548433, 0.43557805,\n", - " 0.30828851, -0.78993368, 0.84360808, 0.93881094, 0.51104885], dtype=float32),\n", - " 'Masters': array([-0.83188468, -1.20755589, 1.29778767, -0.21757448, -0.9422332 ,\n", - " 0.28221262, -0.00386627, -0.95983386, -1.81947958, -0.21439691], dtype=float32),\n", - " 'Preschool': array([ 0.88657302, 0.83354014, 0.39644518, -1.62499583, 0.67740631,\n", - " 0.49476627, 2.9205153 , 1.38263381, 0.0637609 , -0.62043005], dtype=float32),\n", - " 'Prof-school': array([ 0.59643167, -2.98818922, 0.60524243, 0.73828989, 0.18138508,\n", - " -0.22839846, -1.02733207, -1.25480282, 1.0317198 , 0.27936444], dtype=float32),\n", - " 'Some-college': array([ 0.37755287, 0.64273852, 1.17291927, -0.04218093, 0.87608969,\n", - " 1.29980707, -0.35390925, -1.08699441, 0.38861114, -0.59682709], dtype=float32)}" + "{'Bachelors': array([-1.1927266 , 0.13337217, 0.751513 , -0.3854133 , -1.512503 ,\n", + " 0.43075648, 0.03185017, 0.2740599 , -1.3502986 , -0.51524764],\n", + " dtype=float32),\n", + " 'HS-grad': array([ 0.01510752, -0.41036212, -1.2737428 , -0.03190449, 0.30465913,\n", + " -0.4891645 , -0.35087353, 1.7667191 , 0.90333945, -0.42637545],\n", + " dtype=float32),\n", + " '11th': array([-1.3361819 , -1.0304003 , -0.7671982 , 1.1118906 , 0.6290409 ,\n", + " 0.09973534, -0.41261104, -0.79101914, 1.2672484 , 0.7189385 ],\n", + " dtype=float32),\n", + " 'Masters': array([ 0.5837133 , -1.3451334 , 0.9863935 , 0.35932744, -0.13541682,\n", + " 0.34770364, -0.8982047 , 0.4550249 , -1.326133 , -0.08214497],\n", + " dtype=float32),\n", + " '9th': array([ 0.00944321, -0.2883264 , 1.1186845 , 0.16699162, 0.20891678,\n", + " -2.222243 , 0.90257394, -2.499814 , 0.32215422, -0.02830464],\n", + " dtype=float32),\n", + " 'Some-college': array([ 0.11737815, -0.9354352 , -1.6950701 , -0.3879866 , -0.34800476,\n", + " 0.65498114, -1.0632497 , 1.2390918 , -1.3980893 , -1.5068939 ],\n", + " dtype=float32),\n", + " 'Assoc-acdm': array([-0.3248521 , 0.67525595, -0.7607256 , -1.688361 , -0.01024881,\n", + " -0.17185631, -1.5726321 , -0.33589116, -0.6568722 , -0.83356154],\n", + " dtype=float32),\n", + " 'Assoc-voc': array([ 0.11711311, 1.6658193 , 0.2525636 , -1.7053522 , 0.11374688,\n", + " 0.69635576, 0.39209226, 0.55386406, 1.4460421 , -0.4076955 ],\n", + " dtype=float32),\n", + " '7th-8th': array([ 0.8109543 , -0.9696295 , -1.1880634 , -2.673678 , 1.387889 ,\n", + " 0.03207216, 0.28635803, 0.32005164, -0.14126171, -0.12705447],\n", + " dtype=float32),\n", + " 'Doctorate': array([ 2.5456786 , 0.9495662 , -0.65327275, 0.63417935, -1.4665067 ,\n", + " -1.0520831 , -0.8822009 , 1.7168643 , 1.3397688 , 1.0705113 ],\n", + " dtype=float32),\n", + " 'Prof-school': array([-0.3236308 , 0.36975744, 0.79298687, -0.24033554, 0.8012961 ,\n", + " -0.38213903, 0.20259416, -0.30737472, -2.190927 , 0.47054496],\n", + " dtype=float32),\n", + " '5th-6th': array([-0.71498626, -1.3042029 , 0.04956457, -0.20074964, 0.85997975,\n", + " 2.4887364 , 0.9329344 , -0.33221987, -0.37141427, 1.9041626 ],\n", + " dtype=float32),\n", + " '10th': array([-0.5197668 , -1.2800047 , 1.5472891 , -1.141539 , 0.00724531,\n", + " 1.3354197 , 1.4840577 , 0.9995618 , -0.03808165, -1.1237134 ],\n", + " dtype=float32),\n", + " '1st-4th': array([ 1.1701114 , -1.0981313 , -1.5367142 , 0.16519445, -0.0972092 ,\n", + " -0.3711076 , 0.9954778 , -0.94091356, -0.75837976, -1.9332327 ],\n", + " dtype=float32),\n", + " 'Preschool': array([-0.7313262 , -0.56184304, 0.30143896, 0.8417214 , 2.0694172 ,\n", + " -1.2695692 , 1.461705 , -0.6897159 , 1.6769298 , 0.55851436],\n", + " dtype=float32),\n", + " '12th': array([-1.2723062 , -0.27862272, -0.3878713 , -0.9044023 , -0.00804312,\n", + " -1.1498355 , 1.0327121 , 0.29477796, 0.2951289 , 0.96019965],\n", + " dtype=float32)}" ] }, - "execution_count": 14, + "execution_count": 7, "metadata": {}, "output_type": "execute_result" } @@ -465,26 +457,24 @@ }, { "cell_type": "code", - "execution_count": 15, - "metadata": { - "collapsed": false - }, + "execution_count": 8, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "WideDeep (\n", - " (emb_layer_workclass): Embedding(9, 10)\n", - " (emb_layer_education): Embedding(16, 10)\n", + "WideDeep(\n", " (emb_layer_native_country): Embedding(42, 10)\n", " (emb_layer_relationship): Embedding(6, 10)\n", " (emb_layer_occupation): Embedding(15, 10)\n", - " (linear_1): Linear (51 -> 100)\n", - " (linear_1_drop): Dropout (p = 0.5)\n", - " (linear_2): Linear (100 -> 50)\n", - " (linear_2_drop): Dropout (p = 0.2)\n", - " (output): Linear (847 -> 3)\n", + " (emb_layer_education): Embedding(16, 10)\n", + " (emb_layer_workclass): Embedding(9, 10)\n", + " (linear_1): Linear(in_features=51, out_features=100, bias=True)\n", + " (linear_1_drop): Dropout(p=0.5)\n", + " (linear_2): Linear(in_features=100, out_features=50, bias=True)\n", + " (linear_2_drop): Dropout(p=0.2)\n", + " (output): Linear(in_features=847, out_features=3, bias=True)\n", ")\n" ] } @@ -523,33 +513,31 @@ }, { "cell_type": "code", - "execution_count": 16, - "metadata": { - "collapsed": false - }, + "execution_count": 9, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Epoch 1 of 10, Loss: 0.699, accuracy: 0.6737\n", - "Epoch 2 of 10, Loss: 0.822, accuracy: 0.6855\n", - "Epoch 3 of 10, Loss: 0.717, accuracy: 0.6879\n", - "Epoch 4 of 10, Loss: 1.016, accuracy: 0.6931\n", - "Epoch 5 of 10, Loss: 0.842, accuracy: 0.6944\n", - "Epoch 6 of 10, Loss: 0.805, accuracy: 0.6942\n", - "Epoch 7 of 10, Loss: 0.783, accuracy: 0.6966\n", - "Epoch 8 of 10, Loss: 0.859, accuracy: 0.6975\n", - "Epoch 9 of 10, Loss: 0.929, accuracy: 0.6992\n", - "Epoch 10 of 10, Loss: 0.826, accuracy: 0.7006\n", + "Epoch 1 of 10, Loss: 1.131, accuracy: 0.6735\n", + "Epoch 2 of 10, Loss: 0.846, accuracy: 0.6843\n", + "Epoch 3 of 10, Loss: 0.902, accuracy: 0.686\n", + "Epoch 4 of 10, Loss: 0.806, accuracy: 0.691\n", + "Epoch 5 of 10, Loss: 1.015, accuracy: 0.6931\n", + "Epoch 6 of 10, Loss: 0.77, accuracy: 0.694\n", + "Epoch 7 of 10, Loss: 0.868, accuracy: 0.6962\n", + "Epoch 8 of 10, Loss: 0.808, accuracy: 0.6973\n", + "Epoch 9 of 10, Loss: 0.977, accuracy: 0.6972\n", + "Epoch 10 of 10, Loss: 0.851, accuracy: 0.6968\n", "\n", - " [[ 9.99074221e-01 9.25758795e-04 3.93159311e-10]\n", - " [ 2.88534306e-13 1.00000000e+00 1.56172240e-15]\n", - " [ 1.73595769e-08 1.00000000e+00 4.79524920e-10]\n", - " ..., \n", - " [ 8.90251540e-04 9.71086264e-01 2.80234683e-02]\n", - " [ 2.58150152e-07 9.99999106e-01 6.09748270e-07]\n", - " [ 8.45011652e-01 1.54977426e-01 1.09334724e-05]]\n" + " [[9.9808323e-01 1.9167198e-03 1.2708337e-07]\n", + " [1.8705309e-12 1.0000000e+00 1.0048575e-09]\n", + " [2.1682714e-08 9.9999905e-01 9.1604261e-07]\n", + " ...\n", + " [1.0082698e-03 9.6010476e-01 3.8887005e-02]\n", + " [2.4448596e-07 9.9994826e-01 5.1442345e-05]\n", + " [6.8863249e-01 3.0600473e-01 5.3628702e-03]]\n" ] } ], @@ -565,19 +553,17 @@ }, { "cell_type": "code", - "execution_count": 17, - "metadata": { - "collapsed": false - }, + "execution_count": 10, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", - " 0.733689593553\n", + " 0.7318796365288384\n", "\n", - " 0.700402647922\n" + " 0.7006756295639118\n" ] } ], @@ -600,26 +586,24 @@ }, { "cell_type": "code", - "execution_count": 18, - "metadata": { - "collapsed": false - }, + "execution_count": 11, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "WideDeep (\n", - " (emb_layer_workclass): Embedding(9, 10)\n", - " (emb_layer_education): Embedding(16, 10)\n", + "WideDeep(\n", " (emb_layer_native_country): Embedding(42, 10)\n", " (emb_layer_relationship): Embedding(6, 8)\n", " (emb_layer_occupation): Embedding(15, 10)\n", - " (linear_1): Linear (49 -> 100)\n", - " (linear_1_drop): Dropout (p = 0.5)\n", - " (linear_2): Linear (100 -> 50)\n", - " (linear_2_drop): Dropout (p = 0.2)\n", - " (output): Linear (847 -> 1)\n", + " (emb_layer_education): Embedding(16, 10)\n", + " (emb_layer_workclass): Embedding(9, 10)\n", + " (linear_1): Linear(in_features=49, out_features=100, bias=True)\n", + " (linear_1_drop): Dropout(p=0.5)\n", + " (linear_2): Linear(in_features=100, out_features=50, bias=True)\n", + " (linear_2_drop): Dropout(p=0.2)\n", + " (output): Linear(in_features=847, out_features=1, bias=True)\n", ")\n" ] } @@ -652,33 +636,121 @@ }, { "cell_type": "code", - "execution_count": 19, - "metadata": { - "collapsed": false - }, + "execution_count": 12, + "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Epoch 1 of 10, Loss: 151.295\n", - "Epoch 2 of 10, Loss: 108.425\n", - "Epoch 3 of 10, Loss: 82.35\n", - "Epoch 4 of 10, Loss: 36.353\n", - "Epoch 5 of 10, Loss: 50.06\n", - "Epoch 6 of 10, Loss: 147.494\n", - "Epoch 7 of 10, Loss: 176.602\n", - "Epoch 8 of 10, Loss: 167.916\n", - "Epoch 9 of 10, Loss: 40.365\n", - "Epoch 10 of 10, Loss: 107.579\n", + "Epoch 1 of 100, Loss: 246.073\n", + "Epoch 2 of 100, Loss: 94.48\n", + "Epoch 3 of 100, Loss: 126.0\n", + "Epoch 4 of 100, Loss: 115.982\n", + "Epoch 5 of 100, Loss: 133.325\n", + "Epoch 6 of 100, Loss: 158.583\n", + "Epoch 7 of 100, Loss: 54.36\n", + "Epoch 8 of 100, Loss: 79.254\n", + "Epoch 9 of 100, Loss: 182.114\n", + "Epoch 10 of 100, Loss: 51.386\n", + "Epoch 11 of 100, Loss: 104.073\n", + "Epoch 12 of 100, Loss: 143.09\n", + "Epoch 13 of 100, Loss: 70.531\n", + "Epoch 14 of 100, Loss: 97.966\n", + "Epoch 15 of 100, Loss: 59.099\n", + "Epoch 16 of 100, Loss: 94.067\n", + "Epoch 17 of 100, Loss: 183.913\n", + "Epoch 18 of 100, Loss: 47.979\n", + "Epoch 19 of 100, Loss: 179.66\n", + "Epoch 20 of 100, Loss: 137.305\n", + "Epoch 21 of 100, Loss: 127.122\n", + "Epoch 22 of 100, Loss: 229.17\n", + "Epoch 23 of 100, Loss: 151.097\n", + "Epoch 24 of 100, Loss: 161.884\n", + "Epoch 25 of 100, Loss: 128.496\n", + "Epoch 26 of 100, Loss: 157.841\n", + "Epoch 27 of 100, Loss: 369.787\n", + "Epoch 28 of 100, Loss: 94.269\n", + "Epoch 29 of 100, Loss: 134.898\n", + "Epoch 30 of 100, Loss: 74.488\n", + "Epoch 31 of 100, Loss: 214.186\n", + "Epoch 32 of 100, Loss: 112.555\n", + "Epoch 33 of 100, Loss: 78.977\n", + "Epoch 34 of 100, Loss: 134.299\n", + "Epoch 35 of 100, Loss: 35.186\n", + "Epoch 36 of 100, Loss: 76.879\n", + "Epoch 37 of 100, Loss: 195.191\n", + "Epoch 38 of 100, Loss: 113.689\n", + "Epoch 39 of 100, Loss: 104.923\n", + "Epoch 40 of 100, Loss: 155.706\n", + "Epoch 41 of 100, Loss: 115.923\n", + "Epoch 42 of 100, Loss: 67.577\n", + "Epoch 43 of 100, Loss: 105.407\n", + "Epoch 44 of 100, Loss: 125.645\n", + "Epoch 45 of 100, Loss: 133.77\n", + "Epoch 46 of 100, Loss: 149.836\n", + "Epoch 47 of 100, Loss: 173.448\n", + "Epoch 48 of 100, Loss: 110.101\n", + "Epoch 49 of 100, Loss: 138.044\n", + "Epoch 50 of 100, Loss: 153.611\n", + "Epoch 51 of 100, Loss: 61.501\n", + "Epoch 52 of 100, Loss: 71.998\n", + "Epoch 53 of 100, Loss: 188.658\n", + "Epoch 54 of 100, Loss: 80.28\n", + "Epoch 55 of 100, Loss: 95.933\n", + "Epoch 56 of 100, Loss: 136.835\n", + "Epoch 57 of 100, Loss: 129.396\n", + "Epoch 58 of 100, Loss: 66.494\n", + "Epoch 59 of 100, Loss: 111.14\n", + "Epoch 60 of 100, Loss: 176.363\n", + "Epoch 61 of 100, Loss: 75.573\n", + "Epoch 62 of 100, Loss: 53.56\n", + "Epoch 63 of 100, Loss: 89.78\n", + "Epoch 64 of 100, Loss: 93.642\n", + "Epoch 65 of 100, Loss: 105.566\n", + "Epoch 66 of 100, Loss: 159.151\n", + "Epoch 67 of 100, Loss: 89.265\n", + "Epoch 68 of 100, Loss: 121.349\n", + "Epoch 69 of 100, Loss: 98.278\n", + "Epoch 70 of 100, Loss: 249.79\n", + "Epoch 71 of 100, Loss: 111.768\n", + "Epoch 72 of 100, Loss: 137.894\n", + "Epoch 73 of 100, Loss: 131.131\n", + "Epoch 74 of 100, Loss: 83.891\n", + "Epoch 75 of 100, Loss: 99.166\n", + "Epoch 76 of 100, Loss: 219.011\n", + "Epoch 77 of 100, Loss: 62.321\n", + "Epoch 78 of 100, Loss: 109.47\n", + "Epoch 79 of 100, Loss: 122.896\n", + "Epoch 80 of 100, Loss: 58.683\n", + "Epoch 81 of 100, Loss: 199.865\n", + "Epoch 82 of 100, Loss: 122.807\n", + "Epoch 83 of 100, Loss: 138.183\n", + "Epoch 84 of 100, Loss: 152.517\n", + "Epoch 85 of 100, Loss: 34.736\n", + "Epoch 86 of 100, Loss: 34.167\n", + "Epoch 87 of 100, Loss: 152.289\n", + "Epoch 88 of 100, Loss: 58.821\n", + "Epoch 89 of 100, Loss: 70.747\n", + "Epoch 90 of 100, Loss: 259.394\n", + "Epoch 91 of 100, Loss: 79.002\n", + "Epoch 92 of 100, Loss: 80.815\n", + "Epoch 93 of 100, Loss: 152.681\n", + "Epoch 94 of 100, Loss: 103.209\n", + "Epoch 95 of 100, Loss: 75.965\n", + "Epoch 96 of 100, Loss: 67.907\n", + "Epoch 97 of 100, Loss: 96.509\n", + "Epoch 98 of 100, Loss: 182.213\n", + "Epoch 99 of 100, Loss: 55.769\n", + "Epoch 100 of 100, Loss: 115.025\n", "\n", - " RMSE: 11.2378476775\n" + " RMSE: 11.100189228700662\n" ] } ], "source": [ "train_dataset = wd_dataset['train_dataset']\n", - "model.fit(dataset=train_dataset, n_epochs=10, batch_size=64)\n", + "model.fit(dataset=train_dataset, n_epochs=100, batch_size=64)\n", "\n", "test_dataset = wd_dataset['test_dataset']\n", "pred = model.predict(test_dataset)\n", @@ -686,27 +758,34 @@ "from sklearn.metrics import mean_squared_error\n", "print(\"\\n RMSE: {}\".format(np.sqrt(mean_squared_error(pred, test_dataset.labels))))" ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] } ], "metadata": { "kernelspec": { - "display_name": "Python 2", + "display_name": "Python 3", "language": "python", - "name": "python2" + "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", - "version": 2 + "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", - "pygments_lexer": "ipython2", - "version": "2.7.13" + "pygments_lexer": "ipython3", + "version": "3.6.5" } }, "nbformat": 4, - "nbformat_minor": 0 + "nbformat_minor": 1 } diff --git a/wide_deep/data_utils.py b/wide_deep/data_utils.py index 7f8e007..8506e94 100644 --- a/wide_deep/data_utils.py +++ b/wide_deep/data_utils.py @@ -34,10 +34,10 @@ def label_encode(df, cols=None): val_types[c] = df[c].unique() val_to_idx = dict() - for k, v in val_types.iteritems(): + for k, v in val_types.items(): val_to_idx[k] = {o: i for i, o in enumerate(val_types[k])} - for k, v in val_to_idx.iteritems(): + for k, v in val_to_idx.items(): df[k] = df[k].apply(lambda x: v[x]) return val_to_idx, df @@ -106,12 +106,12 @@ def prepare_data(df, wide_cols, crossed_cols, embeddings_cols, continuous_cols, encoding_dict,df_tmp = label_encode(df_tmp) encoding_dict = {k:encoding_dict[k] for k in encoding_dict if k in deep_cols} embeddings_input = [] - for k,v in encoding_dict.iteritems(): + for k,v in encoding_dict.items(): embeddings_input.append((k, len(v), emb_dim[k])) # select the deep_cols and get the column index that will be use later # to slice the tensors - df_deep = df_tmp[deep_cols] + df_deep = df_tmp[deep_cols].copy() deep_column_idx = {k:v for v,k in enumerate(df_deep.columns)} # The continous columns will be concatenated with the embeddings, so you @@ -119,7 +119,7 @@ def prepare_data(df, wide_cols, crossed_cols, embeddings_cols, continuous_cols, if scale: scaler = StandardScaler() for cc in continuous_cols: - df_deep[cc] = scaler.fit_transform(df_deep[cc].values.reshape(-1,1)) + df_deep[cc] = scaler.fit_transform(df_deep[cc].values.reshape(-1,1).astype(float)) # select the wide_cols and one-hot encode those that are categorical df_wide = df_tmp[wide_cols+crossed_columns] diff --git a/wide_deep/torch_model.py b/wide_deep/torch_model.py index 7f83317..55881dc 100644 --- a/wide_deep/torch_model.py +++ b/wide_deep/torch_model.py @@ -8,7 +8,6 @@ import torch.optim as optim from torch.autograd import Variable from torch.utils.data import Dataset, DataLoader - use_cuda = torch.cuda.is_available() @@ -107,7 +106,7 @@ class WideDeep(nn.Module): if method == 'regression': self.activation, self.criterion = None, F.mse_loss if method == 'logistic': - self.activation, self.criterion = F.sigmoid, F.binary_cross_entropy + self.activation, self.criterion = torch.sigmoid, F.binary_cross_entropy if method == 'multiclass': self.activation, self.criterion = F.softmax, F.cross_entropy @@ -153,11 +152,13 @@ class WideDeep(nn.Module): # Deep + Wide sides wide_deep_input = torch.cat([x_deep, X_w.float()], 1) - if not self.activation: out = self.output(wide_deep_input) else: - out = self.activation(self.output(wide_deep_input)) + if (self.activation==F.softmax): + out = self.activation(self.output(wide_deep_input), dim=1) + else: + out = self.activation(self.output(wide_deep_input)) return out @@ -191,8 +192,12 @@ class WideDeep(nn.Module): X_w, X_d, y = X_w.cuda(), X_d.cuda(), y.cuda() self.optimizer.zero_grad() - y_pred = net(X_w, X_d) - loss = self.criterion(y_pred, y) + y_pred = net(X_w, X_d) # [batch_size, 1] + loss = None + if(self.criterion == F.cross_entropy): + loss = self.criterion(y_pred, y) #[batch_size, 1] + else: + loss = self.criterion(y_pred, y.view(-1, 1)) #[batch_size, 1] loss.backward() self.optimizer.step() @@ -202,14 +207,14 @@ class WideDeep(nn.Module): y_pred_cat = (y_pred > 0.5).squeeze(1).float() if self.method == "multiclass": _, y_pred_cat = torch.max(y_pred, 1) - correct+= float((y_pred_cat == y).sum().data[0]) + correct+= float((y_pred_cat == y).sum().item()) if self.method != "regression": print ('Epoch {} of {}, Loss: {}, accuracy: {}'.format(epoch+1, - n_epochs, round(loss.data[0],3), round(correct/total,4))) + n_epochs, round(loss.item(),3), round(correct/total,4))) else: print ('Epoch {} of {}, Loss: {}'.format(epoch+1, n_epochs, - round(loss.data[0],3))) + round(loss.item(),3))) def predict(self, dataset): @@ -293,9 +298,9 @@ class WideDeep(nn.Module): emb_layer = [layer for layer in emb_layers if col_name in layer[0]][0] embeddings = emb_layer[1].cpu().data.numpy() col_label_encoding = self.encoding_dict[col_name] - inv_dict = {v:k for k,v in col_label_encoding.iteritems()} + inv_dict = {v:k for k,v in col_label_encoding.items()} embeddings_dict = {} - for idx,value in inv_dict.iteritems(): + for idx,value in inv_dict.items(): embeddings_dict[value] = embeddings[idx] return embeddings_dict -- GitLab