{"cells":[{"metadata":{},"cell_type":"markdown","source":"My private leaderboard score is 0.9402.\nMy approach:\n\nMean ensemble\n*     Model 1: Efficient B0 - B6 noisy student size 255\n*     Model 2: Efficient B3 - B6 noisy student size 384\n*     Model 3: Efficient B3 - B6 noisy student size 512\n*     Model 4: Multi size Efficient B6 Extra data 2018 + MALIG mean 5 fold \n*     Model 5: Multi size Efficient B6 Dropout connect 0.3 Extra data 2018 + MALIG mean 5 fold\n*     Model 6: Multi size Efficient B6 Shade of gray, Extra data 2018 + MALIG mean 5 fold\n*     Model 7: Multi size Efficient B6 meta input, Extra data 2018 + MALIG mean 5 fold\n*     Model 8: Meta data\n\n=> Private: 0.9402\n     \nThis notebook\n*    Version 2: customize of https://www.kaggle.com/cdeotte/triple-stratified-kfold-with-tfrecords (@cdeotte) *2 + Model 3 *2\n    => Private: 0.9411\n    \n*    Version 6: Remove Model 8(metadata)\n    => Private: 0.9402\n   \n\nThank for @cdeotte's contributes.\n\nThank other public notebooks too. Many ideas are really interesting.\nThey improve my skill.\n\nThank all kagglers.\n","execution_count":null},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nimport os","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sub_path = \"../input/ensemble-melanoma\"\nall_files = os.listdir(sub_path)\nall_files = [f for f in all_files if 'seresnext50 mean tta 0.9252.csv' not in f and 'b6 2019 mean 0.8666.csv' not in f and 'cpu densenet121 0.8845.csv' not in f]\nall_files += ['B3-B6 80 82 size 512.csv', 'triple-stratified-kfold-with-tfrecords 0.9426.csv']\nall_files","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"outs = [pd.read_csv(os.path.join(sub_path, f), index_col=0) for f in all_files]\nconcat_sub = pd.concat(outs, axis=1)\ncols = list(map(lambda x: \"target\" + str(x), range(len(concat_sub.columns))))\nconcat_sub.columns = cols\nconcat_sub.reset_index(inplace=True)\nconcat_sub.head()\nncol = concat_sub.shape[1]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"concat_sub['target'] = concat_sub.iloc[:, 1:ncol].mean(axis=1)\nconcat_sub[['image_name', 'target']].to_csv('submission_mean.csv', index=False, float_format='%.6f')","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}