{"cells":[{"metadata":{},"cell_type":"markdown","source":"# **A nice try**"},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"!pip install --use-feature=2020-resolver https://s3-us-west-2.amazonaws.com/xgboost-nightly-builds/xgboost-1.3.0_SNAPSHOT%2B00b0ad1293b4fa74d6aca5da4e9ab7a9d16777f0-py3-none-manylinux2010_x86_64.whl","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import xgboost as xgb\nxgb.__version__","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport zipfile\nfrom sklearn.model_selection import train_test_split, KFold\nfrom sklearn.metrics import log_loss, roc_auc_score\nfrom sklearn.linear_model import LogisticRegression\nimport gc\nimport os","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission = pd.read_csv('/kaggle/input/dogs-vs-cats-redux-kernels-edition/sample_submission.csv')\nsample_submission.tail()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"TRAIN_FOLDER = '../input/cats-and-dogs-embedded-data/train/train/'\nTEST_FOLDER =  '../input/cats-and-dogs-embedded-data/test/test/'\nIMG_SIZE = 224","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_image_list = np.load('../input/cats-and-dogs-embedded-data/cats_and_dogs_1/train_image_list.npy')\ntest_image_list = np.load('../input/cats-and-dogs-embedded-data/cats_and_dogs_1/test_image_list.npy')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_image_list","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_image_list","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"ids = [int(x.split('.')[0]) for x in test_image_list]\nids[:20]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"target = np.load('../input/cats-and-dogs-embedded-data/cats_and_dogs_1/target.npy')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"target","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_EB7_ns = np.load('../input/cats-and-dogs-embedded-data/cats_and_dogs_4/cats_and_dogs_4/train_EB7_ns.npy')\ntest_EB7_ns = np.load('../input/cats-and-dogs-embedded-data/cats_and_dogs_4/cats_and_dogs_4/test_EB7_ns.npy')\ntrain_EB4_ns = np.load('../input/cats-and-dogs-embedded-data/cats_and_dogs_4/cats_and_dogs_4/train_EB4_ns.npy')\ntest_EB4_ns = np.load('../input/cats-and-dogs-embedded-data/cats_and_dogs_4/cats_and_dogs_4/test_EB4_ns.npy')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_EB7_ns.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"X_train, X_val, y_train, y_val = train_test_split(train_EB7_ns, target, test_size=0.1, random_state=42)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"lr = LogisticRegression(C=0.026, max_iter=10000)\nlr.fit(X_train, y_train)\nval_preds_EB7_lr = lr.predict_proba(X_val)[:,1]\ntest_preds_EB7_lr = lr.predict_proba(test_EB7_ns)[:,1]\nprint(roc_auc_score(y_val, val_preds_EB7_lr))\nprint(log_loss(y_val, val_preds_EB7_lr))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"xgb_params = {\n    'eta': 0.05,\n    'max_depth': 4,\n    'subsample': 0.8,\n    'colsample_bytree': 0.6,\n    'alpha': 0.01,\n    'lambda': 1.00,\n    'gamma' : 0.02,\n    'max_bin': 256,\n    'objective': 'reg:logistic',\n    'eval_metric': 'auc',\n    'verbosity': 0,\n    'tree_method': 'gpu_hist', \n    'predictor': 'gpu_predictor'\n}\n\ndtrain = xgb.DMatrix(X_train, y_train)\ndval = xgb.DMatrix(X_val, y_val)\ndtest = xgb.DMatrix(test_EB7_ns)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"del train_EB7_ns, test_EB7_ns, X_train, X_val\ngc.collect()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model = xgb.train(xgb_params, dtrain, num_boost_round=100)\nval_preds_xgb_1 = model.predict(dval)\ntest_preds_xgb_1 = model.predict(dtest)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"roc_auc_score(y_val, val_preds_xgb_1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"log_loss(y_val, val_preds_xgb_1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"roc_auc_score(y_val, 0.5*val_preds_xgb_1+0.5*val_preds_EB7_lr)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"log_loss(y_val, 0.5*val_preds_xgb_1+0.5*val_preds_EB7_lr)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_preds_xgb_1.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"X_train, X_val, y_train, y_val = train_test_split(train_EB4_ns, target, test_size=0.1, random_state=42)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"lr = LogisticRegression(C=0.019, max_iter=10000)\nlr.fit(X_train, y_train)\nval_preds_EB4_lr = lr.predict_proba(X_val)[:,1]\ntest_preds_EB4_lr = lr.predict_proba(test_EB4_ns)[:,1]\nprint(roc_auc_score(y_val, val_preds_EB4_lr))\nprint(log_loss(y_val, val_preds_EB4_lr))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"xgb_params = {\n    'eta': 0.05,\n    'max_depth': 4,\n    'subsample': 0.85,\n    'colsample_bytree': 0.6,\n    'alpha': 0.01,\n    'lambda': 1.00,\n    'gamma' : 0.02,\n    'max_bin': 256,\n    'objective': 'reg:logistic',\n    'eval_metric': 'auc',\n    'verbosity': 0,\n    'tree_method': 'gpu_hist', \n    'predictor': 'gpu_predictor'\n}\n\ndtrain = xgb.DMatrix(X_train, y_train)\ndval = xgb.DMatrix(X_val, y_val)\ndtest = xgb.DMatrix(test_EB4_ns)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"del train_EB4_ns, test_EB4_ns, X_train, X_val\ngc.collect()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"val_preds_xgb_2 = 0\ntest_preds_xgb_2 = 0\n\nn_seed = 10\n\nfor j in range(n_seed):\n    xgb_params['seed'] = 3*j**2+154\n    model = xgb.train(xgb_params, dtrain, num_boost_round=100)\n    val_preds_xgb_2 += model.predict(dval)/n_seed\n    test_preds_xgb_2 += model.predict(dtest)/n_seed","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = test_preds_xgb_1\nsample_submission.to_csv('submission_xgb_1.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = test_preds_xgb_2\nsample_submission.to_csv('submission_xgb_2.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = 0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr\nsample_submission.to_csv('submission_0.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = (0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)**0.99\nsample_submission.to_csv('submission_0_b.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = (0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)**0.95\nsample_submission.to_csv('submission_0_c.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = 0.47*test_preds_xgb_1+0.53*test_preds_EB7_lr\nsample_submission.to_csv('submission_1.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = 0.52*test_preds_xgb_1+0.48*test_preds_EB7_lr\nsample_submission.to_csv('submission_2.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = 0.55*test_preds_xgb_1+0.45*test_preds_EB7_lr\nsample_submission.to_csv('submission_3.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = 0.5*test_preds_xgb_2+0.5*test_preds_EB4_lr\nsample_submission.to_csv('submission_EB4_0.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = np.clip(1.01*(0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)-0.01*(0.5*test_preds_xgb_2+0.5*test_preds_EB4_lr), 0, 1)\nsample_submission.to_csv('submission_EB7_EB4_0.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = np.clip(1.05*(0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)-0.05*(0.5*test_preds_xgb_2+0.5*test_preds_EB4_lr), 0, 1)\nsample_submission.to_csv('submission_EB7_EB4_1.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = np.clip(1.08*(0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)-0.08*(0.5*test_preds_xgb_2+0.5*test_preds_EB4_lr), 0, 1)\nsample_submission.to_csv('submission_EB7_EB4_2.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = np.clip(1.08*(0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)-0.08*(0.5*test_preds_xgb_2+0.5*test_preds_EB4_lr), 0, 1)\nsample_submission.to_csv('submission_EB7_EB4_2.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsubs = np.clip(1.1*(0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)-0.1*(0.5*test_preds_xgb_2+0.5*test_preds_EB4_lr), 0, 1)\nsample_submission['label'] = subs\nsample_submission.to_csv('submission_EB7_EB4_3.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsubs = np.clip(1.1*(0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)-0.1*(0.5*test_preds_xgb_2+0.5*test_preds_EB4_lr), 0, 1)\nsubs[subs > 0.9976] = 1\nsample_submission['label'] = subs\nsample_submission.to_csv('submission_EB7_EB4_3_b.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsubs = np.clip(1.1*(0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)-0.1*(0.5*test_preds_xgb_2+0.5*test_preds_EB4_lr), 0, 1)\nsubs[subs > 0.9973] = 1\nsample_submission['label'] = subs\nsample_submission.to_csv('submission_EB7_EB4_3_c.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsubs = np.clip(1.1*(0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)-0.1*(0.5*test_preds_xgb_2+0.5*test_preds_EB4_lr), 0, 1)\nsubs[subs > 0.9973] = 1\nsubs[subs <  0.0018] = 0\nsample_submission['label'] = subs\nsample_submission.to_csv('submission_EB7_EB4_3_d.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsample_submission['label'] = np.clip(1.12*(0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)-0.12*(0.5*test_preds_xgb_2+0.5*test_preds_EB4_lr), 0, 1)\nsample_submission.to_csv('submission_EB7_EB4_4.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_submission['id'] = ids\nsubs = np.clip(1.12*(0.5*test_preds_xgb_1+0.5*test_preds_EB7_lr)-0.12*(0.5*test_preds_xgb_2+0.5*test_preds_EB4_lr), 0, 1)\n\nsample_submission['label'] = subs\nsubs[subs > 0.9974] = 1\nsubs[subs <  0.0018] = 0\nsample_submission.to_csv('submission_EB7_EB4_4_b.csv', index=False)\nsample_submission.head()","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat":4,"nbformat_minor":4}