playground-s3-e22-very-simple-approach.ipynb

{"cells":[{"source":"<a href=\"https://www.kaggle.com/code/yaaangzhou/pg-s3-e22-very-simple-approach?scriptVersionId=145414756\" target=\"_blank\"><img align=\"left\" alt=\"Kaggle\" title=\"Open in Kaggle\" src=\"https://kaggle.com/static/images/open-in-kaggle.svg\"></a>","metadata":{},"cell_type":"markdown"},{"cell_type":"markdown","id":"31c885ca","metadata":{"papermill":{"duration":0.003439,"end_time":"2023-10-06T04:01:08.687162","exception":false,"start_time":"2023-10-06T04:01:08.683723","status":"completed"},"tags":[]},"source":["Based on public kernals. thanks to all for your efforts.\n","\n","EDA and FSS come from this [notebook](https://www.kaggle.com/code/yaaangzhou/playground-s3-e22-eda-modeling)."]},{"cell_type":"code","execution_count":1,"id":"cc97daf7","metadata":{"execution":{"iopub.execute_input":"2023-10-06T04:01:08.697746Z","iopub.status.busy":"2023-10-06T04:01:08.697034Z","iopub.status.idle":"2023-10-06T04:01:12.115273Z","shell.execute_reply":"2023-10-06T04:01:12.114289Z"},"papermill":{"duration":3.425306,"end_time":"2023-10-06T04:01:12.117663","exception":false,"start_time":"2023-10-06T04:01:08.692357","status":"completed"},"tags":[]},"outputs":[],"source":["import numpy as np\n","import pandas as pd\n","\n","from sklearn.preprocessing import LabelEncoder\n","\n","# Models\n","from lightgbm import LGBMClassifier\n","from sklearn.ensemble import HistGradientBoostingClassifier\n","\n","import warnings\n","warnings.filterwarnings('ignore')"]},{"cell_type":"markdown","id":"fdec92d7","metadata":{"papermill":{"duration":0.002638,"end_time":"2023-10-06T04:01:12.123469","exception":false,"start_time":"2023-10-06T04:01:12.120831","status":"completed"},"tags":[]},"source":["# Prepare data"]},{"cell_type":"code","execution_count":2,"id":"ab6bce74","metadata":{"execution":{"iopub.execute_input":"2023-10-06T04:01:12.131296Z","iopub.status.busy":"2023-10-06T04:01:12.130913Z","iopub.status.idle":"2023-10-06T04:01:12.332995Z","shell.execute_reply":"2023-10-06T04:01:12.331881Z"},"papermill":{"duration":0.209161,"end_time":"2023-10-06T04:01:12.335491","exception":false,"start_time":"2023-10-06T04:01:12.12633","status":"completed"},"tags":[]},"outputs":[],"source":["train = pd.read_csv('/kaggle/input/playground-series-s3e22/train.csv')\n","test = pd.read_csv('/kaggle/input/playground-series-s3e22/test.csv')\n","sample_submission = pd.read_csv('/kaggle/input/playground-series-s3e22/sample_submission.csv')\n","origin = pd.read_csv('/kaggle/input/horse-survival-dataset/horse.csv')\n","\n","train[\"is_generated\"] = 1\n","test[\"is_generated\"] = 1\n","origin[\"is_generated\"] = 0\n","\n","# Drop column id\n","train.drop('id',axis=1,inplace=True)\n","test.drop('id',axis=1,inplace=True)\n","\n","train_total = pd.concat([train, origin], ignore_index=True)\n","train_total.drop_duplicates(inplace=True)\n","\n","total = pd.concat([train_total, test], ignore_index=True)\n","\n","target = 'outcome'\n","\n","total[target] = total[target].map({'died':0,'euthanized':1,'lived':2})"]},{"cell_type":"code","execution_count":3,"id":"1f2d0b94","metadata":{"execution":{"iopub.execute_input":"2023-10-06T04:01:12.343221Z","iopub.status.busy":"2023-10-06T04:01:12.342753Z","iopub.status.idle":"2023-10-06T04:01:12.386549Z","shell.execute_reply":"2023-10-06T04:01:12.385216Z"},"papermill":{"duration":0.050362,"end_time":"2023-10-06T04:01:12.388864","exception":false,"start_time":"2023-10-06T04:01:12.338502","status":"completed"},"tags":[]},"outputs":[],"source":["def preprocessing(df, le_cols, ohe_cols):\n","    \n","    # Label Encoding for binary cols\n","    le = LabelEncoder()    \n","    for col in le_cols:\n","        df[col] = le.fit_transform(df[col])\n","    \n","    # OneHot Encoding for category cols\n","    df = pd.get_dummies(df, columns = ohe_cols)\n","    \n","    df[\"pain\"] = df[\"pain\"].replace('slight', 'moderate')\n","    df[\"peristalsis\"] = df[\"peristalsis\"].replace('distend_small', 'normal')\n","    df[\"rectal_exam_feces\"] = df[\"rectal_exam_feces\"].replace('serosanguious', 'absent')\n","    df[\"nasogastric_reflux\"] = df[\"nasogastric_reflux\"].replace('slight', 'none')\n","        \n","    df[\"temp_of_extremities\"] = df[\"temp_of_extremities\"].fillna(\"normal\").map({'cold': 0, 'cool': 1, 'normal': 2, 'warm': 3})\n","    df[\"peripheral_pulse\"] = df[\"peripheral_pulse\"].fillna(\"normal\").map({'absent': 0, 'reduced': 1, 'normal': 2, 'increased': 3})\n","    df[\"capillary_refill_time\"] = df[\"capillary_refill_time\"].fillna(\"3\").map({'less_3_sec': 0, '3': 1, 'more_3_sec': 2})\n","    df[\"pain\"] = df[\"pain\"].fillna(\"depressed\").map({'alert': 0, 'depressed': 1, 'moderate': 2, 'mild_pain': 3, 'severe_pain': 4, 'extreme_pain': 5})\n","    df[\"peristalsis\"] = df[\"peristalsis\"].fillna(\"hypomotile\").map({'hypermotile': 0, 'normal': 1, 'hypomotile': 2, 'absent': 3})\n","    df[\"abdominal_distention\"] = df[\"abdominal_distention\"].fillna(\"none\").map({'none': 0, 'slight': 1, 'moderate': 2, 'severe': 3})\n","    df[\"nasogastric_tube\"] = df[\"nasogastric_tube\"].fillna(\"none\").map({'none': 0, 'slight': 1, 'significant': 2})\n","    df[\"nasogastric_reflux\"] = df[\"nasogastric_reflux\"].fillna(\"none\").map({'less_1_liter': 0, 'none': 1, 'more_1_liter': 2})\n","    df[\"rectal_exam_feces\"] = df[\"rectal_exam_feces\"].fillna(\"absent\").map({'absent': 0, 'decreased': 1, 'normal': 2, 'increased': 3})\n","    df[\"abdomen\"] = df[\"abdomen\"].fillna(\"distend_small\").map({'normal': 0, 'other': 1, 'firm': 2,'distend_small': 3, 'distend_large': 4})\n","    df[\"abdomo_appearance\"] = df[\"abdomo_appearance\"].fillna(\"serosanguious\").map({'clear': 0, 'cloudy': 1, 'serosanguious': 2})\n","    \n","    # Imputer \n","    cols_with_nan = df.drop(target,axis=1).columns[df.drop(target,axis=1).isna().any()].tolist()\n","    for feature in cols_with_nan:\n","        df[feature].fillna(df[feature].mode()[0], inplace=True)\n","     \n","    return df  \n","\n","total = preprocessing(total, le_cols = [\"surgery\", \"age\", \"surgical_lesion\", \"cp_data\"], ohe_cols = [\"mucous_membrane\"])"]},{"cell_type":"code","execution_count":4,"id":"2345c253","metadata":{"execution":{"iopub.execute_input":"2023-10-06T04:01:12.396653Z","iopub.status.busy":"2023-10-06T04:01:12.396299Z","iopub.status.idle":"2023-10-06T04:01:12.402514Z","shell.execute_reply":"2023-10-06T04:01:12.401348Z"},"papermill":{"duration":0.012905,"end_time":"2023-10-06T04:01:12.404862","exception":false,"start_time":"2023-10-06T04:01:12.391957","status":"completed"},"tags":[]},"outputs":[],"source":["del_features = ['lesion_2', 'lesion_3', 'mucous_membrane_dark_cyanotic']\n","\n","total = total.drop(del_features,axis=1)"]},{"cell_type":"code","execution_count":5,"id":"116404aa","metadata":{"execution":{"iopub.execute_input":"2023-10-06T04:01:12.4131Z","iopub.status.busy":"2023-10-06T04:01:12.41232Z","iopub.status.idle":"2023-10-06T04:01:12.4243Z","shell.execute_reply":"2023-10-06T04:01:12.422984Z"},"papermill":{"duration":0.01862,"end_time":"2023-10-06T04:01:12.426581","exception":false,"start_time":"2023-10-06T04:01:12.407961","status":"completed"},"tags":[]},"outputs":[],"source":["df_train = total[total[target].notna()]\n","df_test = total[total[target].isna()]\n","df_test.drop(target,axis=1,inplace=True)"]},{"cell_type":"code","execution_count":6,"id":"acea48d6","metadata":{"execution":{"iopub.execute_input":"2023-10-06T04:01:12.434755Z","iopub.status.busy":"2023-10-06T04:01:12.434347Z","iopub.status.idle":"2023-10-06T04:01:12.446905Z","shell.execute_reply":"2023-10-06T04:01:12.445877Z"},"papermill":{"duration":0.019754,"end_time":"2023-10-06T04:01:12.449421","exception":false,"start_time":"2023-10-06T04:01:12.429667","status":"completed"},"tags":[]},"outputs":[],"source":["full_features = df_test.columns.tolist()\n","bin_features = df_test.select_dtypes('bool').columns\n","\n","df_train[bin_features] = df_train[bin_features].astype('int64')\n","df_test[bin_features] = df_test[bin_features].astype('int64')"]},{"cell_type":"markdown","id":"018417cd","metadata":{"papermill":{"duration":0.002789,"end_time":"2023-10-06T04:01:12.455538","exception":false,"start_time":"2023-10-06T04:01:12.452749","status":"completed"},"tags":[]},"source":["# Training model"]},{"cell_type":"code","execution_count":7,"id":"3d67018d","metadata":{"execution":{"iopub.execute_input":"2023-10-06T04:01:12.463282Z","iopub.status.busy":"2023-10-06T04:01:12.462913Z","iopub.status.idle":"2023-10-06T04:01:12.90714Z","shell.execute_reply":"2023-10-06T04:01:12.906153Z"},"papermill":{"duration":0.451142,"end_time":"2023-10-06T04:01:12.909619","exception":false,"start_time":"2023-10-06T04:01:12.458477","status":"completed"},"tags":[]},"outputs":[],"source":["hist_md = HistGradientBoostingClassifier(\n","    max_depth=4,          \n","    max_iter=80,         \n","    learning_rate=0.1,     \n","    random_state=42,   \n","    scoring='f1_micro',          \n","    max_leaf_nodes = 21,\n","    l2_regularization = 0.1,\n",").fit(df_train.drop(target,axis=1), df_train[target])\n","\n","hist_pred = hist_md.predict(df_test)"]},{"cell_type":"markdown","id":"f050b72f","metadata":{"papermill":{"duration":0.002972,"end_time":"2023-10-06T04:01:12.916033","exception":false,"start_time":"2023-10-06T04:01:12.913061","status":"completed"},"tags":[]},"source":["# Submission"]},{"cell_type":"code","execution_count":8,"id":"16fe9a07","metadata":{"execution":{"iopub.execute_input":"2023-10-06T04:01:12.924257Z","iopub.status.busy":"2023-10-06T04:01:12.923129Z","iopub.status.idle":"2023-10-06T04:01:12.934493Z","shell.execute_reply":"2023-10-06T04:01:12.933306Z"},"papermill":{"duration":0.017848,"end_time":"2023-10-06T04:01:12.936899","exception":false,"start_time":"2023-10-06T04:01:12.919051","status":"completed"},"tags":[]},"outputs":[],"source":["sample_submission['outcome'] = hist_pred\n","sample_submission['outcome'] = sample_submission['outcome'].map({0:'died',1:'euthanized',2:'lived'})\n","sample_submission.to_csv('submission.csv',index=False)"]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.10.12"},"papermill":{"default_parameters":{},"duration":8.274634,"end_time":"2023-10-06T04:01:13.661743","environment_variables":{},"exception":null,"input_path":"__notebook__.ipynb","output_path":"__notebook__.ipynb","parameters":{},"start_time":"2023-10-06T04:01:05.387109","version":"2.4.0"}},"nbformat":4,"nbformat_minor":5}