logreg_tuning.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"orig_nbformat":4,"colab":{"name":"logreg.ipynb","provenance":[],"collapsed_sections":[]},"language_info":{"name":"python"},"kernelspec":{"name":"python3","display_name":"Python 3"}},"cells":[{"cell_type":"code","metadata":{"id":"_6JdaoewkOIO","executionInfo":{"status":"ok","timestamp":1634769041217,"user_tz":-180,"elapsed":296,"user":{"displayName":"Nikita Pachshenko","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gg9JYlOY0OxBo1_zatjPUgEu20AoaKaEVnYbMuC=s64","userId":"00713447482913267047"}}},"source":["import numpy as np\n","import pandas as pd\n","import matplotlib.pyplot as plt\n","\n","from sklearn.metrics import accuracy_score, recall_score, precision_score\n","from sklearn.metrics import f1_score, make_scorer\n","from sklearn.model_selection import cross_validate, StratifiedKFold, train_test_split\n","\n","from sklearn.linear_model import LogisticRegression\n","from xgboost import XGBClassifier"],"execution_count":15,"outputs":[]},{"cell_type":"code","metadata":{"id":"21nB38IokOIZ","executionInfo":{"status":"ok","timestamp":1634768205416,"user_tz":-180,"elapsed":309,"user":{"displayName":"Nikita Pachshenko","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gg9JYlOY0OxBo1_zatjPUgEu20AoaKaEVnYbMuC=s64","userId":"00713447482913267047"}}},"source":["from sklearn.model_selection import GridSearchCV"],"execution_count":2,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"2IPb7dQJkPsD","executionInfo":{"status":"ok","timestamp":1634768205421,"user_tz":-180,"elapsed":13,"user":{"displayName":"Nikita Pachshenko","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gg9JYlOY0OxBo1_zatjPUgEu20AoaKaEVnYbMuC=s64","userId":"00713447482913267047"}},"outputId":"1d0f41ad-5d5e-4ee3-f2b2-d81feaf1a2c2"},"source":["from google.colab import drive\n","drive.mount('/content/drive')"],"execution_count":3,"outputs":[{"output_type":"stream","name":"stdout","text":["Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"]}]},{"cell_type":"code","metadata":{"id":"dBYdApQLkOIa","executionInfo":{"status":"ok","timestamp":1634768205687,"user_tz":-180,"elapsed":274,"user":{"displayName":"Nikita Pachshenko","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gg9JYlOY0OxBo1_zatjPUgEu20AoaKaEVnYbMuC=s64","userId":"00713447482913267047"}}},"source":["X = np.load('drive/MyDrive/DATA/data_2/X.npy').astype(np.int8)\n","y = np.load('drive/MyDrive/DATA/data_2/y.npy').astype(np.int8)"],"execution_count":4,"outputs":[]},{"cell_type":"code","metadata":{"id":"Kc0Ovz48kOIc","executionInfo":{"status":"ok","timestamp":1634768205688,"user_tz":-180,"elapsed":13,"user":{"displayName":"Nikita Pachshenko","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gg9JYlOY0OxBo1_zatjPUgEu20AoaKaEVnYbMuC=s64","userId":"00713447482913267047"}}},"source":["scoring = {'accuracy': 'accuracy',\n","           'F1-score': make_scorer(f1_score, average='micro'),\n","           }"],"execution_count":5,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"qP44JMiLkOId","executionInfo":{"status":"ok","timestamp":1634768564736,"user_tz":-180,"elapsed":359059,"user":{"displayName":"Nikita Pachshenko","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gg9JYlOY0OxBo1_zatjPUgEu20AoaKaEVnYbMuC=s64","userId":"00713447482913267047"}},"outputId":"267c2f5b-514b-45c8-806c-00f5c7a86f61"},"source":["parameters = {'penalty': ['l1', 'l2'],\n","              'C':[0.01, 0.1, 1, 10, 100],\n","              'class_weight': [None, 'balanced']}\n","logreg = LogisticRegression(solver='liblinear')\n","gs = GridSearchCV(logreg, parameters, scoring, cv=5, verbose=4, refit='F1-score', n_jobs=-1)\n","gs.fit(X, y)"],"execution_count":6,"outputs":[{"output_type":"stream","name":"stdout","text":["Fitting 5 folds for each of 20 candidates, totalling 100 fits\n"]},{"output_type":"stream","name":"stderr","text":["[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n","[Parallel(n_jobs=-1)]: Done  21 tasks      | elapsed:   54.4s\n","[Parallel(n_jobs=-1)]: Done  94 tasks      | elapsed:  5.4min\n","[Parallel(n_jobs=-1)]: Done 100 out of 100 | elapsed:  5.9min finished\n"]},{"output_type":"execute_result","data":{"text/plain":["GridSearchCV(cv=5, error_score=nan,\n","             estimator=LogisticRegression(C=1.0, class_weight=None, dual=False,\n","                                          fit_intercept=True,\n","                                          intercept_scaling=1, l1_ratio=None,\n","                                          max_iter=100, multi_class='auto',\n","                                          n_jobs=None, penalty='l2',\n","                                          random_state=None, solver='liblinear',\n","                                          tol=0.0001, verbose=0,\n","                                          warm_start=False),\n","             iid='deprecated', n_jobs=-1,\n","             param_grid={'C': [0.01, 0.1, 1, 10, 100],\n","                         'class_weight': [None, 'balanced'],\n","                         'penalty': ['l1', 'l2']},\n","             pre_dispatch='2*n_jobs', refit='F1-score',\n","             return_train_score=False,\n","             scoring={'F1-score': make_scorer(f1_score, average=micro),\n","                      'accuracy': 'accuracy'},\n","             verbose=4)"]},"metadata":{},"execution_count":6}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Kzo6ZyJKw3gN","executionInfo":{"status":"ok","timestamp":1634769384558,"user_tz":-180,"elapsed":17,"user":{"displayName":"Nikita Pachshenko","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gg9JYlOY0OxBo1_zatjPUgEu20AoaKaEVnYbMuC=s64","userId":"00713447482913267047"}},"outputId":"73f939c3-79cf-4468-8367-6079ff24ebb4"},"source":["gs.best_score_"],"execution_count":26,"outputs":[{"output_type":"execute_result","data":{"text/plain":["0.7532603285216526"]},"metadata":{},"execution_count":26}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"jTIO2gEG13_H","executionInfo":{"status":"ok","timestamp":1634769289603,"user_tz":-180,"elapsed":280,"user":{"displayName":"Nikita Pachshenko","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gg9JYlOY0OxBo1_zatjPUgEu20AoaKaEVnYbMuC=s64","userId":"00713447482913267047"}},"outputId":"32befaeb-0fca-4b42-9cc6-dd8fdd67fa99"},"source":["y = np.where(y == 1, 0, 1)\n","np.unique(y)"],"execution_count":20,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([0, 1])"]},"metadata":{},"execution_count":20}]},{"cell_type":"code","metadata":{"id":"Bdyv5jql0q_1","executionInfo":{"status":"ok","timestamp":1634769290034,"user_tz":-180,"elapsed":7,"user":{"displayName":"Nikita Pachshenko","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gg9JYlOY0OxBo1_zatjPUgEu20AoaKaEVnYbMuC=s64","userId":"00713447482913267047"}}},"source":["scoring = {'accuracy': 'accuracy',\n","           'precision': 'precision',\n","           'recall': 'recall',\n","           'F1-score': 'f1',\n","           'recall -': make_scorer(recall_score, pos_label=0),\n","           'precision -': make_scorer(precision_score, pos_label=0, zero_division=0),\n","           }\n","\n","scores_df = pd.DataFrame(index=['LogisticRegression'], columns=list(scoring.keys()))"],"execution_count":21,"outputs":[]},{"cell_type":"code","metadata":{"id":"FiAmATrj0_zR","executionInfo":{"status":"ok","timestamp":1634769309345,"user_tz":-180,"elapsed":19008,"user":{"displayName":"Nikita Pachshenko","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gg9JYlOY0OxBo1_zatjPUgEu20AoaKaEVnYbMuC=s64","userId":"00713447482913267047"}}},"source":["clf = gs.best_estimator_\n","scores = cross_validate(clf, X, y, cv=5, scoring=scoring, n_jobs=-1)\n","mean_scores = pd.DataFrame(scores).apply(np.mean)"],"execution_count":22,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":81},"id":"HCHJCS9F1M2Q","executionInfo":{"status":"ok","timestamp":1634769309355,"user_tz":-180,"elapsed":23,"user":{"displayName":"Nikita Pachshenko","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14Gg9JYlOY0OxBo1_zatjPUgEu20AoaKaEVnYbMuC=s64","userId":"00713447482913267047"}},"outputId":"cfc9daa7-8604-4921-ce6b-15a2514f1c17"},"source":["scores_df.loc[type(clf).__name__] = mean_scores.values[2:]\n","scores_df"],"execution_count":23,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>accuracy</th>\n","      <th>precision</th>\n","      <th>recall</th>\n","      <th>F1-score</th>\n","      <th>recall -</th>\n","      <th>precision -</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>LogisticRegression</th>\n","      <td>0.75326</td>\n","      <td>0.752859</td>\n","      <td>1</td>\n","      <td>0.859006</td>\n","      <td>0.00655738</td>\n","      <td>0.4</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["                   accuracy precision recall  F1-score    recall - precision -\n","LogisticRegression  0.75326  0.752859      1  0.859006  0.00655738         0.4"]},"metadata":{},"execution_count":23}]}]}