diff --git "a/src/test.ipynb" "b/src/test.ipynb" --- "a/src/test.ipynb" +++ "b/src/test.ipynb" @@ -2,7 +2,7 @@ "cells": [ { "cell_type": "code", - "execution_count": 170, + "execution_count": 25, "metadata": { "collapsed": true, "pycharm": { @@ -24,7 +24,7 @@ }, { "cell_type": "code", - "execution_count": 157, + "execution_count": 26, "outputs": [], "source": [ "def get_data() -> tuple[Any, Any, Any]:\n", @@ -50,34 +50,25 @@ }, { "cell_type": "code", - "execution_count": 171, + "execution_count": 27, "outputs": [], "source": [ - "data, target, treatment = get_data()\n", - "uplift = [1 for _ in data.index]" - ], - "metadata": { - "collapsed": false, - "pycharm": { - "name": "#%%\n" - } - } - }, - { - "cell_type": "code", - "execution_count": 172, - "outputs": [ - { - "data": { - "text/plain": "0.04515395574087702" - }, - "execution_count": 172, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "weighted_average_uplift(target, uplift, treatment)" + "dataset, target, treatment = get_data()\n", + "\n", + "data_train_index = pd.read_csv('../data/data_train_index.csv')\n", + "data_test_index = pd.read_csv('../data/data_test_index.csv')\n", + "treatment_train_index = pd.read_csv('../data/treatment_train_index.csv')\n", + "treatment_test_index = pd.read_csv('../data/treatment_test_index.csv')\n", + "target_train_index = pd.read_csv('../data/target_train_index.csv')\n", + "target_test_index = pd.read_csv('../data/target_test_index.csv')\n", + "\n", + "# фиксируем выборки, чтобы результат работы ML был предсказуем\n", + "data_train = dataset.loc[data_train_index['0']]\n", + "data_test = dataset.loc[data_test_index['0']]\n", + "treatment_train = treatment.loc[treatment_train_index['0']]\n", + "treatment_test = treatment.loc[treatment_test_index['0']]\n", + "target_train = target.loc[target_train_index['0']]\n", + "target_test = target.loc[target_test_index['0']]" ], "metadata": { "collapsed": false, @@ -88,13 +79,9 @@ }, { "cell_type": "code", - "execution_count": 159, + "execution_count": 28, "outputs": [], "source": [ - "X_train, X_val, y_train, y_val, trmnt_train, trmnt_val = train_test_split(\n", - " data, target, treatment, test_size=0.5, random_state=42\n", - ")\n", - "\n", "models_results = {\n", " 'approach': [],\n", " 'uplift@30%': []\n", @@ -108,37 +95,54 @@ } }, { - "cell_type": "code", - "execution_count": 162, - "outputs": [], + "cell_type": "markdown", "source": [ - "new_val = X_val.loc[X_val[X_val['newbie'] == 1].index]\n", - "new_y = y_val.loc[new_val.index]\n", - "new_trmt = trmnt_val.loc[new_val.index]\n", - "uplift = [np.random.random() for _ in new_val.index]" + "## Single model" ], "metadata": { "collapsed": false, "pycharm": { - "name": "#%%\n" + "name": "#%% md\n" } } }, { "cell_type": "code", - "execution_count": 163, + "execution_count": 44, "outputs": [ { "data": { - "text/plain": "0.07056871131907891" + "text/plain": "
", + "image/png": "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\n" }, - "execution_count": 163, - "metadata": {}, - "output_type": "execute_result" + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" } ], "source": [ - "uplift_at_k(y_true=new_y, uplift=uplift, treatment=new_trmt, strategy='by_group', k=0.3)" + "# cbr = CatBoostClassifier(iterations=500, task_type=\"GPU\", random_state=42, silent=True)\n", + "cbr = CatBoostClassifier()\n", + "cbr.load_model('sm_cbc.cbm')\n", + "sm = SoloModel(cbr)\n", + "# sm.estimator.load_model('model.cbm')\n", + "# sm = sm.fit(data_train, treatment_train, target_train, estimator_fit_params={'cat_features': ['womens', 'mens','channel', 'zip_code', 'history_segment', 'newbie']})\n", + "\n", + "uplift_sm = sm.predict(data_test)\n", + "\n", + "sm_score = uplift_at_k(y_true=target_test, uplift=uplift_sm, treatment=treatment_test, strategy='by_group', k=0.3)\n", + "\n", + "models_results['approach'].append('SoloModel')\n", + "models_results['uplift@30%'].append(sm_score)\n", + "\n", + "# Получим условные вероятности выполнения целевого действия при взаимодействии для каждого объекта\n", + "sm_trmnt_preds = sm.trmnt_preds_\n", + "# И условные вероятности выполнения целевого действия без взаимодействия для каждого объекта\n", + "sm_ctrl_preds = sm.ctrl_preds_\n", + "\n", + "# Отрисуем распределения вероятностей и их разность (uplift)\n", + "plot_uplift_preds(trmnt_preds=sm_trmnt_preds, ctrl_preds=sm_ctrl_preds);" ], "metadata": { "collapsed": false, @@ -149,20 +153,10 @@ }, { "cell_type": "code", - "execution_count": 165, - "outputs": [ - { - "data": { - "text/plain": " n_treatment n_control response_rate_treatment \\\npercentile \n0-10 549 515 0.149362 \n10-20 553 511 0.113924 \n20-30 511 553 0.138943 \n30-40 546 518 0.130037 \n40-50 544 520 0.156250 \n50-60 542 521 0.153137 \n60-70 518 545 0.106178 \n70-80 529 534 0.130435 \n80-90 528 535 0.128788 \n90-100 539 524 0.116883 \n\n response_rate_control uplift \npercentile \n0-10 0.054369 0.094994 \n10-20 0.066536 0.047388 \n20-30 0.070524 0.068419 \n30-40 0.090734 0.039303 \n40-50 0.080769 0.075481 \n50-60 0.078695 0.074442 \n60-70 0.056881 0.049297 \n70-80 0.071161 0.059274 \n80-90 0.069159 0.059629 \n90-100 0.085878 0.031005 ", - "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_treatmentn_controlresponse_rate_treatmentresponse_rate_controluplift
percentile
0-105495150.1493620.0543690.094994
10-205535110.1139240.0665360.047388
20-305115530.1389430.0705240.068419
30-405465180.1300370.0907340.039303
40-505445200.1562500.0807690.075481
50-605425210.1531370.0786950.074442
60-705185450.1061780.0568810.049297
70-805295340.1304350.0711610.059274
80-905285350.1287880.0691590.059629
90-1005395240.1168830.0858780.031005
\n
" - }, - "execution_count": 165, - "metadata": {}, - "output_type": "execute_result" - } - ], + "execution_count": 30, + "outputs": [], "source": [ - "uplift_by_percentile(new_y, uplift, new_trmt)" + "sm.estimator.save_model('models/sm_cbc.cbm')" ], "metadata": { "collapsed": false, @@ -171,26 +165,30 @@ } } }, - { - "cell_type": "markdown", - "source": [ - "## Single model" - ], - "metadata": { - "collapsed": false, - "pycharm": { - "name": "#%% md\n" - } - } - }, { "cell_type": "code", - "execution_count": 120, + "execution_count": 31, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "P:\\uplift_lab\\venv\\lib\\site-packages\\numpy\\core\\fromnumeric.py:3156: 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", + " return asarray(a).ndim\n" + ] + }, { "data": { - "text/plain": "
", - "image/png": "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\n" + "text/plain": "" + }, + "execution_count": 31, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "text/plain": "
", + "image/png": "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\n" }, "metadata": { "needs_background": "light" @@ -199,23 +197,9 @@ } ], "source": [ - "sm = SoloModel(CatBoostClassifier(iterations=100, task_type=\"GPU\", random_state=42, silent=True))\n", - "sm = sm.fit(X_train, y_train, trmnt_train, estimator_fit_params={'cat_features': ['history_segment', 'zip_code', 'channel']})\n", - "\n", - "uplift_sm = sm.predict(X_val)\n", - "\n", - "sm_score = uplift_at_k(y_true=y_val, uplift=uplift_sm, treatment=trmnt_val, strategy='by_group', k=0.3)\n", - "\n", - "models_results['approach'].append('SoloModel')\n", - "models_results['uplift@30%'].append(sm_score)\n", - "\n", - "# Получим условные вероятности выполнения целевого действия при взаимодействии для каждого объекта\n", - "sm_trmnt_preds = sm.trmnt_preds_\n", - "# И условные вероятности выполнения целевого действия без взаимодействия для каждого объекта\n", - "sm_ctrl_preds = sm.ctrl_preds_\n", + "from sklift.viz import plot_uplift_by_percentile\n", "\n", - "# Отрисуем распределения вероятностей и их разность (uplift)\n", - "plot_uplift_preds(trmnt_preds=sm_trmnt_preds, ctrl_preds=sm_ctrl_preds);" + "plot_uplift_by_percentile(target_test, uplift_sm, treatment_test)" ], "metadata": { "collapsed": false, @@ -226,14 +210,14 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 32, "outputs": [ { "data": { - "text/plain": " feature_name feature_score\n0 newbie 16.159375\n1 history 15.380477\n2 recency 15.057106\n3 womens 12.918455\n4 treatment 12.083391\n5 channel 9.437116\n6 zip_code 9.357430\n7 mens 6.807001\n8 history_segment 2.799648", - "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
feature_namefeature_score
0newbie16.159375
1history15.380477
2recency15.057106
3womens12.918455
4treatment12.083391
5channel9.437116
6zip_code9.357430
7mens6.807001
8history_segment2.799648
\n
" + "text/plain": " feature_name feature_score\n0 treatment 23.090004\n1 channel 19.284459\n2 zip_code 15.911562\n3 history_segment 13.699184\n4 history 13.131164\n5 recency 7.301571\n6 mens 3.429339\n7 womens 3.394266\n8 newbie 0.758451", + "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
feature_namefeature_score
0treatment23.090004
1channel19.284459
2zip_code15.911562
3history_segment13.699184
4history13.131164
5recency7.301571
6mens3.429339
7womens3.394266
8newbie0.758451
\n
" }, - "execution_count": 14, + "execution_count": 32, "metadata": {}, "output_type": "execute_result" } @@ -267,21 +251,35 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 33, "outputs": [], "source": [ "from sklift.models import ClassTransformation\n", "\n", "\n", - "ct = ClassTransformation(CatBoostClassifier(iterations=1000, task_type='GPU', random_state=42, silent=True))\n", - "ct = ct.fit(X_train, y_train, trmnt_train, estimator_fit_params={'cat_features': ['history_segment', 'zip_code', 'channel']})\n", + "ct = ClassTransformation(CatBoostClassifier(iterations=500, task_type='GPU', random_state=42, silent=True))\n", + "ct.fit(data_train, target_train, treatment_train, estimator_fit_params={'cat_features': ['womens', 'mens','channel', 'zip_code', 'history_segment', 'newbie']})\n", "\n", - "uplift_ct = ct.predict(X_val)\n", + "uplift_ct = ct.predict(data_test)\n", "\n", - "ct_score = uplift_at_k(y_true=y_val, uplift=uplift_ct, treatment=trmnt_val, strategy='by_group', k=0.3)\n", + "ct_score = uplift_at_k(y_true=target_test, uplift=uplift_ct, treatment=treatment_test, strategy='by_group', k=0.3)\n", "\n", "models_results['approach'].append('ClassTransformation')\n", - "models_results['uplift@30%'].append(ct_score)" + "models_results['uplift@30%'].append(ct_score)\n" + ], + "metadata": { + "collapsed": false, + "pycharm": { + "name": "#%%\n" + } + } + }, + { + "cell_type": "code", + "execution_count": 34, + "outputs": [], + "source": [ + "ct.estimator.save_model('models/ct_cbc.cbm')" ], "metadata": { "collapsed": false, @@ -304,12 +302,12 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 35, "outputs": [ { "data": { "text/plain": "
", - "image/png": "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\n" + "image/png": "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\n" }, "metadata": { "needs_background": "light" @@ -322,19 +320,19 @@ "\n", "\n", "tm = TwoModels(\n", - " estimator_trmnt=CatBoostClassifier(iterations=1000, task_type='GPU', random_state=42, silent=True),\n", - " estimator_ctrl=CatBoostClassifier(iterations=1000, task_type='GPU', random_state=42, silent=True),\n", + " estimator_trmnt=CatBoostClassifier(iterations=500, task_type='GPU', random_state=42, silent=True),\n", + " estimator_ctrl=CatBoostClassifier(iterations=500, task_type='GPU', random_state=42, silent=True),\n", " method='vanilla'\n", ")\n", - "tm = tm.fit(\n", - " X_train, y_train, trmnt_train,\n", - " estimator_trmnt_fit_params={'cat_features': ['history_segment', 'zip_code', 'channel']},\n", - " estimator_ctrl_fit_params={'cat_features': ['history_segment', 'zip_code', 'channel']}\n", + "tm.fit(\n", + " data_train, target_train, treatment_train,\n", + " estimator_trmnt_fit_params={'cat_features': ['womens', 'mens','channel', 'zip_code', 'history_segment', 'newbie']},\n", + " estimator_ctrl_fit_params={'cat_features': ['womens', 'mens','channel', 'zip_code', 'history_segment', 'newbie']}\n", ")\n", "\n", - "uplift_tm = tm.predict(X_val)\n", + "uplift_tm = tm.predict(data_test)\n", "\n", - "tm_score = uplift_at_k(y_true=y_val, uplift=uplift_tm, treatment=trmnt_val, strategy='by_group', k=0.3)\n", + "tm_score = uplift_at_k(y_true=target_test, uplift=uplift_tm, treatment=treatment_test, strategy='by_group', k=0.3)\n", "\n", "models_results['approach'].append('TwoModels')\n", "models_results['uplift@30%'].append(tm_score)\n", @@ -348,6 +346,21 @@ } } }, + { + "cell_type": "code", + "execution_count": 36, + "outputs": [], + "source": [ + "tm.estimator_ctrl.save_model('models/tm_ctrl_cbc.cbm')\n", + "tm.estimator_trmnt.save_model('models/tm_trmnt_cbc.cbm')" + ], + "metadata": { + "collapsed": false, + "pycharm": { + "name": "#%%\n" + } + } + }, { "cell_type": "markdown", "source": [ @@ -362,12 +375,12 @@ }, { "cell_type": "code", - "execution_count": 36, + "execution_count": 37, "outputs": [ { "data": { "text/plain": "
", - "image/png": "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\n" + "image/png": "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\n" }, "metadata": { "needs_background": "light" @@ -377,19 +390,100 @@ ], "source": [ "tm_ctrl = TwoModels(\n", - " estimator_trmnt=CatBoostClassifier(iterations=1000, task_type='GPU', random_state=42, silent=True),\n", - " estimator_ctrl=CatBoostClassifier(iterations=1000, task_type='GPU', random_state=42, silent=True),\n", + " estimator_trmnt=CatBoostClassifier(iterations=500, task_type='GPU', random_state=42, silent=True),\n", + " estimator_ctrl=CatBoostClassifier(iterations=500, task_type='GPU', random_state=42, silent=True),\n", + " method='ddr_control'\n", + ")\n", + "\n", + "tm_ctrl = tm_ctrl.fit(\n", + " data_train, target_train, treatment_train,\n", + " estimator_trmnt_fit_params={'cat_features': ['womens', 'mens','channel', 'zip_code', 'history_segment', 'newbie']},\n", + " estimator_ctrl_fit_params={'cat_features': ['womens', 'mens','channel', 'zip_code', 'history_segment', 'newbie']}\n", + ")\n", + "\n", + "uplift_tm_ctrl = tm_ctrl.predict(data_test)\n", + "\n", + "tm_ctrl_score = uplift_at_k(y_true=target_test, uplift=uplift_tm_ctrl, treatment=treatment_test, strategy='by_group', k=0.3)\n", + "\n", + "models_results['approach'].append('TwoModels_ddr_control')\n", + "models_results['uplift@30%'].append(tm_ctrl_score)\n", + "\n", + "plot_uplift_preds(trmnt_preds=tm_ctrl.trmnt_preds_, ctrl_preds=tm_ctrl.ctrl_preds_);" + ], + "metadata": { + "collapsed": false, + "pycharm": { + "name": "#%%\n" + } + } + }, + { + "cell_type": "code", + "execution_count": 38, + "outputs": [], + "source": [ + "tm.estimator_ctrl.save_model('models/tm_dependend_ctrl_cbc.cbm')\n", + "tm.estimator_trmnt.save_model('models/tm_dependend_trmnt_cbc.cbm')" + ], + "metadata": { + "collapsed": false, + "pycharm": { + "name": "#%%\n" + } + } + }, + { + "cell_type": "code", + "execution_count": 52, + "outputs": [ + { + "ename": "ValueError", + "evalue": "Shape mismatch: if categories is an array, it has to be of shape (n_features,).", + "output_type": "error", + "traceback": [ + "\u001B[1;31m---------------------------------------------------------------------------\u001B[0m", + "\u001B[1;31mValueError\u001B[0m Traceback (most recent call last)", + "Input \u001B[1;32mIn [52]\u001B[0m, in \u001B[0;36m\u001B[1;34m()\u001B[0m\n\u001B[0;32m 10\u001B[0m pipeline_trtmnt \u001B[38;5;241m=\u001B[39m make_pipeline(\n\u001B[0;32m 11\u001B[0m OrdinalEncoder(categories\u001B[38;5;241m=\u001B[39m[\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mchannel\u001B[39m\u001B[38;5;124m'\u001B[39m, \u001B[38;5;124m'\u001B[39m\u001B[38;5;124mzip_code\u001B[39m\u001B[38;5;124m'\u001B[39m, \u001B[38;5;124m'\u001B[39m\u001B[38;5;124mhistory_segment\u001B[39m\u001B[38;5;124m'\u001B[39m]),\n\u001B[0;32m 12\u001B[0m RandomForestClassifier(n_estimators\u001B[38;5;241m=\u001B[39m\u001B[38;5;241m100\u001B[39m, max_depth\u001B[38;5;241m=\u001B[39m\u001B[38;5;241m5\u001B[39m, random_state\u001B[38;5;241m=\u001B[39m\u001B[38;5;241m42\u001B[39m)\n\u001B[0;32m 13\u001B[0m )\n\u001B[0;32m 15\u001B[0m tm_ctrl \u001B[38;5;241m=\u001B[39m TwoModels(\n\u001B[0;32m 16\u001B[0m estimator_trmnt\u001B[38;5;241m=\u001B[39mpipeline_ctrl,\n\u001B[0;32m 17\u001B[0m estimator_ctrl\u001B[38;5;241m=\u001B[39mpipeline_trtmnt,\n\u001B[0;32m 18\u001B[0m method\u001B[38;5;241m=\u001B[39m\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mddr_control\u001B[39m\u001B[38;5;124m'\u001B[39m\n\u001B[0;32m 19\u001B[0m )\n\u001B[1;32m---> 21\u001B[0m tm_ctrl \u001B[38;5;241m=\u001B[39m \u001B[43mtm_ctrl\u001B[49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43mfit\u001B[49m\u001B[43m(\u001B[49m\n\u001B[0;32m 22\u001B[0m \u001B[43m \u001B[49m\u001B[43mdata_train\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43mtarget_train\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43mtreatment_train\u001B[49m\n\u001B[0;32m 23\u001B[0m \u001B[43m)\u001B[49m\n\u001B[0;32m 25\u001B[0m uplift_tm_ctrl \u001B[38;5;241m=\u001B[39m tm_ctrl\u001B[38;5;241m.\u001B[39mpredict(data_test)\n\u001B[0;32m 27\u001B[0m tm_ctrl_score \u001B[38;5;241m=\u001B[39m uplift_at_k(y_true\u001B[38;5;241m=\u001B[39mtarget_test, uplift\u001B[38;5;241m=\u001B[39muplift_tm_ctrl, treatment\u001B[38;5;241m=\u001B[39mtreatment_test, strategy\u001B[38;5;241m=\u001B[39m\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mby_group\u001B[39m\u001B[38;5;124m'\u001B[39m, k\u001B[38;5;241m=\u001B[39m\u001B[38;5;241m0.3\u001B[39m)\n", + "File \u001B[1;32mP:\\uplift_lab\\venv\\lib\\site-packages\\sklift\\models\\models.py:401\u001B[0m, in \u001B[0;36mTwoModels.fit\u001B[1;34m(self, X, y, treatment, estimator_trmnt_fit_params, estimator_ctrl_fit_params)\u001B[0m\n\u001B[0;32m 396\u001B[0m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mestimator_trmnt\u001B[38;5;241m.\u001B[39mfit(\n\u001B[0;32m 397\u001B[0m X_trmnt, y_trmnt, \u001B[38;5;241m*\u001B[39m\u001B[38;5;241m*\u001B[39mestimator_trmnt_fit_params\n\u001B[0;32m 398\u001B[0m )\n\u001B[0;32m 400\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mmethod \u001B[38;5;241m==\u001B[39m \u001B[38;5;124m'\u001B[39m\u001B[38;5;124mddr_control\u001B[39m\u001B[38;5;124m'\u001B[39m:\n\u001B[1;32m--> 401\u001B[0m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mestimator_ctrl\u001B[38;5;241m.\u001B[39mfit(\n\u001B[0;32m 402\u001B[0m X_ctrl, y_ctrl, \u001B[38;5;241m*\u001B[39m\u001B[38;5;241m*\u001B[39mestimator_ctrl_fit_params\n\u001B[0;32m 403\u001B[0m )\n\u001B[0;32m 404\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39m_type_of_target \u001B[38;5;241m==\u001B[39m \u001B[38;5;124m'\u001B[39m\u001B[38;5;124mbinary\u001B[39m\u001B[38;5;124m'\u001B[39m:\n\u001B[0;32m 405\u001B[0m ddr_control \u001B[38;5;241m=\u001B[39m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mestimator_ctrl\u001B[38;5;241m.\u001B[39mpredict_proba(X_trmnt)[:, \u001B[38;5;241m1\u001B[39m]\n", + "File \u001B[1;32mP:\\uplift_lab\\venv\\lib\\site-packages\\sklearn\\pipeline.py:378\u001B[0m, in \u001B[0;36mPipeline.fit\u001B[1;34m(self, X, y, **fit_params)\u001B[0m\n\u001B[0;32m 352\u001B[0m \u001B[38;5;124;03m\"\"\"Fit the model.\u001B[39;00m\n\u001B[0;32m 353\u001B[0m \n\u001B[0;32m 354\u001B[0m \u001B[38;5;124;03mFit all the transformers one after the other and transform the\u001B[39;00m\n\u001B[1;32m (...)\u001B[0m\n\u001B[0;32m 375\u001B[0m \u001B[38;5;124;03m Pipeline with fitted steps.\u001B[39;00m\n\u001B[0;32m 376\u001B[0m \u001B[38;5;124;03m\"\"\"\u001B[39;00m\n\u001B[0;32m 377\u001B[0m fit_params_steps \u001B[38;5;241m=\u001B[39m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39m_check_fit_params(\u001B[38;5;241m*\u001B[39m\u001B[38;5;241m*\u001B[39mfit_params)\n\u001B[1;32m--> 378\u001B[0m Xt \u001B[38;5;241m=\u001B[39m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39m_fit(X, y, \u001B[38;5;241m*\u001B[39m\u001B[38;5;241m*\u001B[39mfit_params_steps)\n\u001B[0;32m 379\u001B[0m \u001B[38;5;28;01mwith\u001B[39;00m _print_elapsed_time(\u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mPipeline\u001B[39m\u001B[38;5;124m\"\u001B[39m, \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39m_log_message(\u001B[38;5;28mlen\u001B[39m(\u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39msteps) \u001B[38;5;241m-\u001B[39m \u001B[38;5;241m1\u001B[39m)):\n\u001B[0;32m 380\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39m_final_estimator \u001B[38;5;241m!=\u001B[39m \u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mpassthrough\u001B[39m\u001B[38;5;124m\"\u001B[39m:\n", + "File \u001B[1;32mP:\\uplift_lab\\venv\\lib\\site-packages\\sklearn\\pipeline.py:336\u001B[0m, in \u001B[0;36mPipeline._fit\u001B[1;34m(self, X, y, **fit_params_steps)\u001B[0m\n\u001B[0;32m 334\u001B[0m cloned_transformer \u001B[38;5;241m=\u001B[39m clone(transformer)\n\u001B[0;32m 335\u001B[0m \u001B[38;5;66;03m# Fit or load from cache the current transformer\u001B[39;00m\n\u001B[1;32m--> 336\u001B[0m X, fitted_transformer \u001B[38;5;241m=\u001B[39m fit_transform_one_cached(\n\u001B[0;32m 337\u001B[0m cloned_transformer,\n\u001B[0;32m 338\u001B[0m X,\n\u001B[0;32m 339\u001B[0m y,\n\u001B[0;32m 340\u001B[0m \u001B[38;5;28;01mNone\u001B[39;00m,\n\u001B[0;32m 341\u001B[0m message_clsname\u001B[38;5;241m=\u001B[39m\u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mPipeline\u001B[39m\u001B[38;5;124m\"\u001B[39m,\n\u001B[0;32m 342\u001B[0m message\u001B[38;5;241m=\u001B[39m\u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39m_log_message(step_idx),\n\u001B[0;32m 343\u001B[0m \u001B[38;5;241m*\u001B[39m\u001B[38;5;241m*\u001B[39mfit_params_steps[name],\n\u001B[0;32m 344\u001B[0m )\n\u001B[0;32m 345\u001B[0m \u001B[38;5;66;03m# Replace the transformer of the step with the fitted\u001B[39;00m\n\u001B[0;32m 346\u001B[0m \u001B[38;5;66;03m# transformer. This is necessary when loading the transformer\u001B[39;00m\n\u001B[0;32m 347\u001B[0m \u001B[38;5;66;03m# from the cache.\u001B[39;00m\n\u001B[0;32m 348\u001B[0m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39msteps[step_idx] \u001B[38;5;241m=\u001B[39m (name, fitted_transformer)\n", + "File \u001B[1;32mP:\\uplift_lab\\venv\\lib\\site-packages\\joblib\\memory.py:349\u001B[0m, in \u001B[0;36mNotMemorizedFunc.__call__\u001B[1;34m(self, *args, **kwargs)\u001B[0m\n\u001B[0;32m 348\u001B[0m \u001B[38;5;28;01mdef\u001B[39;00m \u001B[38;5;21m__call__\u001B[39m(\u001B[38;5;28mself\u001B[39m, \u001B[38;5;241m*\u001B[39margs, \u001B[38;5;241m*\u001B[39m\u001B[38;5;241m*\u001B[39mkwargs):\n\u001B[1;32m--> 349\u001B[0m \u001B[38;5;28;01mreturn\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mfunc(\u001B[38;5;241m*\u001B[39margs, \u001B[38;5;241m*\u001B[39m\u001B[38;5;241m*\u001B[39mkwargs)\n", + "File \u001B[1;32mP:\\uplift_lab\\venv\\lib\\site-packages\\sklearn\\pipeline.py:870\u001B[0m, in \u001B[0;36m_fit_transform_one\u001B[1;34m(transformer, X, y, weight, message_clsname, message, **fit_params)\u001B[0m\n\u001B[0;32m 868\u001B[0m \u001B[38;5;28;01mwith\u001B[39;00m _print_elapsed_time(message_clsname, message):\n\u001B[0;32m 869\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m \u001B[38;5;28mhasattr\u001B[39m(transformer, \u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mfit_transform\u001B[39m\u001B[38;5;124m\"\u001B[39m):\n\u001B[1;32m--> 870\u001B[0m res \u001B[38;5;241m=\u001B[39m transformer\u001B[38;5;241m.\u001B[39mfit_transform(X, y, \u001B[38;5;241m*\u001B[39m\u001B[38;5;241m*\u001B[39mfit_params)\n\u001B[0;32m 871\u001B[0m \u001B[38;5;28;01melse\u001B[39;00m:\n\u001B[0;32m 872\u001B[0m res \u001B[38;5;241m=\u001B[39m transformer\u001B[38;5;241m.\u001B[39mfit(X, y, \u001B[38;5;241m*\u001B[39m\u001B[38;5;241m*\u001B[39mfit_params)\u001B[38;5;241m.\u001B[39mtransform(X)\n", + "File \u001B[1;32mP:\\uplift_lab\\venv\\lib\\site-packages\\sklearn\\base.py:870\u001B[0m, in \u001B[0;36mTransformerMixin.fit_transform\u001B[1;34m(self, X, y, **fit_params)\u001B[0m\n\u001B[0;32m 867\u001B[0m \u001B[38;5;28;01mreturn\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mfit(X, \u001B[38;5;241m*\u001B[39m\u001B[38;5;241m*\u001B[39mfit_params)\u001B[38;5;241m.\u001B[39mtransform(X)\n\u001B[0;32m 868\u001B[0m \u001B[38;5;28;01melse\u001B[39;00m:\n\u001B[0;32m 869\u001B[0m \u001B[38;5;66;03m# fit method of arity 2 (supervised transformation)\u001B[39;00m\n\u001B[1;32m--> 870\u001B[0m \u001B[38;5;28;01mreturn\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mfit(X, y, \u001B[38;5;241m*\u001B[39m\u001B[38;5;241m*\u001B[39mfit_params)\u001B[38;5;241m.\u001B[39mtransform(X)\n", + "File \u001B[1;32mP:\\uplift_lab\\venv\\lib\\site-packages\\sklearn\\preprocessing\\_encoders.py:1294\u001B[0m, in \u001B[0;36mOrdinalEncoder.fit\u001B[1;34m(self, X, y)\u001B[0m\n\u001B[0;32m 1287\u001B[0m \u001B[38;5;28;01mraise\u001B[39;00m \u001B[38;5;167;01mTypeError\u001B[39;00m(\n\u001B[0;32m 1288\u001B[0m \u001B[38;5;124m\"\u001B[39m\u001B[38;5;124munknown_value should only be set when \u001B[39m\u001B[38;5;124m\"\u001B[39m\n\u001B[0;32m 1289\u001B[0m \u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mhandle_unknown is \u001B[39m\u001B[38;5;124m'\u001B[39m\u001B[38;5;124muse_encoded_value\u001B[39m\u001B[38;5;124m'\u001B[39m\u001B[38;5;124m, \u001B[39m\u001B[38;5;124m\"\u001B[39m\n\u001B[0;32m 1290\u001B[0m \u001B[38;5;124mf\u001B[39m\u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mgot \u001B[39m\u001B[38;5;132;01m{\u001B[39;00m\u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39munknown_value\u001B[38;5;132;01m}\u001B[39;00m\u001B[38;5;124m.\u001B[39m\u001B[38;5;124m\"\u001B[39m\n\u001B[0;32m 1291\u001B[0m )\n\u001B[0;32m 1293\u001B[0m \u001B[38;5;66;03m# `_fit` will only raise an error when `self.handle_unknown=\"error\"`\u001B[39;00m\n\u001B[1;32m-> 1294\u001B[0m \u001B[38;5;28;43mself\u001B[39;49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43m_fit\u001B[49m\u001B[43m(\u001B[49m\u001B[43mX\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43mhandle_unknown\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[38;5;28;43mself\u001B[39;49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43mhandle_unknown\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43mforce_all_finite\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[38;5;124;43m\"\u001B[39;49m\u001B[38;5;124;43mallow-nan\u001B[39;49m\u001B[38;5;124;43m\"\u001B[39;49m\u001B[43m)\u001B[49m\n\u001B[0;32m 1296\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mhandle_unknown \u001B[38;5;241m==\u001B[39m \u001B[38;5;124m\"\u001B[39m\u001B[38;5;124muse_encoded_value\u001B[39m\u001B[38;5;124m\"\u001B[39m:\n\u001B[0;32m 1297\u001B[0m \u001B[38;5;28;01mfor\u001B[39;00m feature_cats \u001B[38;5;129;01min\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mcategories_:\n", + "File \u001B[1;32mP:\\uplift_lab\\venv\\lib\\site-packages\\sklearn\\preprocessing\\_encoders.py:87\u001B[0m, in \u001B[0;36m_BaseEncoder._fit\u001B[1;34m(self, X, handle_unknown, force_all_finite, return_counts)\u001B[0m\n\u001B[0;32m 85\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mcategories \u001B[38;5;241m!=\u001B[39m \u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mauto\u001B[39m\u001B[38;5;124m\"\u001B[39m:\n\u001B[0;32m 86\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m \u001B[38;5;28mlen\u001B[39m(\u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mcategories) \u001B[38;5;241m!=\u001B[39m n_features:\n\u001B[1;32m---> 87\u001B[0m \u001B[38;5;28;01mraise\u001B[39;00m \u001B[38;5;167;01mValueError\u001B[39;00m(\n\u001B[0;32m 88\u001B[0m \u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mShape mismatch: if categories is an array,\u001B[39m\u001B[38;5;124m\"\u001B[39m\n\u001B[0;32m 89\u001B[0m \u001B[38;5;124m\"\u001B[39m\u001B[38;5;124m it has to be of shape (n_features,).\u001B[39m\u001B[38;5;124m\"\u001B[39m\n\u001B[0;32m 90\u001B[0m )\n\u001B[0;32m 92\u001B[0m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mcategories_ \u001B[38;5;241m=\u001B[39m []\n\u001B[0;32m 93\u001B[0m category_counts \u001B[38;5;241m=\u001B[39m []\n", + "\u001B[1;31mValueError\u001B[0m: Shape mismatch: if categories is an array, it has to be of shape (n_features,)." + ] + } + ], + "source": [ + "from sklearn.ensemble import RandomForestClassifier\n", + "from sklearn.preprocessing import OrdinalEncoder\n", + "from sklearn.pipeline import make_pipeline\n", + "\n", + "pipeline_ctrl = make_pipeline(\n", + " OrdinalEncoder(categories=['channel', 'zip_code', 'history_segment']),\n", + " RandomForestClassifier(n_estimators=100, max_depth=5, random_state=42)\n", + ")\n", + "\n", + "pipeline_trtmnt = make_pipeline(\n", + " OrdinalEncoder(categories=['channel', 'zip_code', 'history_segment']),\n", + " RandomForestClassifier(n_estimators=100, max_depth=5, random_state=42)\n", + ")\n", + "\n", + "tm_ctrl = TwoModels(\n", + " estimator_trmnt=pipeline_ctrl,\n", + " estimator_ctrl=pipeline_trtmnt,\n", " method='ddr_control'\n", ")\n", + "\n", "tm_ctrl = tm_ctrl.fit(\n", - " X_train, y_train, trmnt_train,\n", - " estimator_trmnt_fit_params={'cat_features': ['gender']},\n", - " estimator_ctrl_fit_params={'cat_features': ['gender']}\n", + " data_train, target_train, treatment_train\n", ")\n", "\n", - "uplift_tm_ctrl = tm_ctrl.predict(X_val)\n", + "uplift_tm_ctrl = tm_ctrl.predict(data_test)\n", "\n", - "tm_ctrl_score = uplift_at_k(y_true=y_val, uplift=uplift_tm_ctrl, treatment=trmnt_val, strategy='by_group', k=0.3)\n", + "tm_ctrl_score = uplift_at_k(y_true=target_test, uplift=uplift_tm_ctrl, treatment=treatment_test, strategy='by_group', k=0.3)\n", "\n", "models_results['approach'].append('TwoModels_ddr_control')\n", "models_results['uplift@30%'].append(tm_ctrl_score)\n", @@ -405,14 +499,14 @@ }, { "cell_type": "code", - "execution_count": 47, + "execution_count": 39, "outputs": [ { "data": { - "text/plain": " 0 1 2 3 4 5 6 \\\n0 0.116088 0.269822 0.049340 0.038516 0.143292 0.047044 0.018165 \n1 0.045442 0.041342 -0.018277 0.012866 0.028861 0.007020 -0.001542 \n\n 7 8 9 ... 206099 206100 206101 206102 \\\n0 0.086771 0.137409 0.029504 ... 0.191889 0.044686 0.069574 0.170429 \n1 0.047911 -0.013122 -0.004837 ... 0.076940 -0.005309 0.033985 -0.097014 \n\n 206103 206104 206105 206106 206107 206108 \n0 0.384708 0.022230 0.020183 0.043997 0.283808 0.053526 \n1 -0.093607 0.002797 0.000479 0.014596 -0.098411 0.024334 \n\n[2 rows x 206109 columns]", - "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
0123456789...206099206100206101206102206103206104206105206106206107206108
00.1160880.2698220.0493400.0385160.1432920.0470440.0181650.0867710.1374090.029504...0.1918890.0446860.0695740.1704290.3847080.0222300.0201830.0439970.2838080.053526
10.0454420.041342-0.0182770.0128660.0288610.007020-0.0015420.047911-0.013122-0.004837...0.076940-0.0053090.033985-0.097014-0.0936070.0027970.0004790.014596-0.0984110.024334
\n

2 rows × 206109 columns

\n
" + "text/plain": " 0 1 2 3 4 5 6 \\\n0 0.187557 0.105281 0.161889 0.140160 0.468606 0.172560 0.130849 \n1 0.036023 0.028304 0.045443 0.069228 0.181322 0.113093 0.053770 \n\n 7 8 9 ... 21337 21338 21339 21340 \\\n0 0.079300 0.039731 0.114872 ... 0.091512 0.254877 0.163009 0.089335 \n1 0.000879 0.005583 0.026389 ... 0.045846 0.089228 0.013675 -0.014415 \n\n 21341 21342 21343 21344 21345 21346 \n0 0.200761 0.215388 0.142818 0.231624 0.232737 0.143152 \n1 0.111710 0.087185 0.035382 -0.003007 -0.011135 0.048037 \n\n[2 rows x 21347 columns]", + "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
0123456789...21337213382133921340213412134221343213442134521346
00.1875570.1052810.1618890.1401600.4686060.1725600.1308490.0793000.0397310.114872...0.0915120.2548770.1630090.0893350.2007610.2153880.1428180.2316240.2327370.143152
10.0360230.0283040.0454430.0692280.1813220.1130930.0537700.0008790.0055830.026389...0.0458460.0892280.013675-0.0144150.1117100.0871850.035382-0.003007-0.0111350.048037
\n

2 rows × 21347 columns

\n
" }, - "execution_count": 47, + "execution_count": 39, "metadata": {}, "output_type": "execute_result" } @@ -441,14 +535,14 @@ }, { "cell_type": "code", - "execution_count": 37, + "execution_count": 40, "outputs": [ { "data": { - "text/plain": " approach uplift@30%\n3 ClassTransformation 0.031702\n4 TwoModels 0.016004\n1 SoloModel 0.013953\n2 SoloModel 0.013411\n5 TwoModels_ddr_control 0.011537\n0 SoloModel 0.006042", - "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
approachuplift@30%
3ClassTransformation0.031702
4TwoModels0.016004
1SoloModel0.013953
2SoloModel0.013411
5TwoModels_ddr_control0.011537
0SoloModel0.006042
\n
" + "text/plain": " approach uplift@30%\n0 SoloModel 0.077602\n3 TwoModels_ddr_control 0.073417\n2 TwoModels 0.073161\n1 ClassTransformation 0.066895", + "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
approachuplift@30%
0SoloModel0.077602
3TwoModels_ddr_control0.073417
2TwoModels0.073161
1ClassTransformation0.066895
\n
" }, - "execution_count": 37, + "execution_count": 40, "metadata": {}, "output_type": "execute_result" } @@ -465,20 +559,20 @@ }, { "cell_type": "code", - "execution_count": 39, + "execution_count": 41, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "C:\\Users\\HardWorkingStation\\PyProjects\\uplift_lab\\venv\\lib\\site-packages\\numpy\\core\\fromnumeric.py:3156: 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", + "P:\\uplift_lab\\venv\\lib\\site-packages\\numpy\\core\\fromnumeric.py:3156: 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", " return asarray(a).ndim\n" ] }, { "data": { "text/plain": "
", - "image/png": "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\n" + "image/png": "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\n" }, "metadata": { "needs_background": "light" @@ -490,30 +584,7 @@ "from sklift.viz import plot_uplift_by_percentile\n", "\n", "# line plot\n", - "plot_uplift_by_percentile(y_val, uplift_ct, trmnt_val, strategy='overall', kind='bar');" - ], - "metadata": { - "collapsed": false, - "pycharm": { - "name": "#%%\n" - } - } - }, - { - "cell_type": "code", - "execution_count": 48, - "outputs": [ - { - "data": { - "text/plain": "sklearn.utils._bunch.Bunch" - }, - "execution_count": 48, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "type(dataset)" + "plot_uplift_by_percentile(target_test, uplift_ct, treatment_test, strategy='overall', kind='bar');" ], "metadata": { "collapsed": false,