{"cells":[{"metadata":{},"cell_type":"markdown","source":"**I am new to this idea of blending different outputs to achieve better prediction results. In this notebook I tried to apply this idea.**","execution_count":null},{"metadata":{},"cell_type":"markdown","source":"**I forked this notebook from : https://www.kaggle.com/servietsky/dark-magic-blending-0-9603**\n\n\nI blended outputs from :\n\n<a href=\"https://www.kaggle.com/solomonk/minmax-ensemble-0-9526-lb\">MinMax Ensemble (0.9526 LB)</a>\n\n<a href=\"https://www.kaggle.com/truonghoang/stacking-ensemble-on-my-submissions\">Stacking, Ensemble on my submissions</a>\n\n<a href=\"https://www.kaggle.com/datafan07/analysis-of-melanoma-metadata-and-effnet-ensemble\">Analysis of Melanoma Metadata and EffNet Ensemble</a>\n\n<a href=\"https://www.kaggle.com/datafan07/eda-modelling-of-the-external-data-inc-ensemble\">EDA&Modelling of the External Data Inc. Ensemble</a>\n\n<a href=\"https://www.kaggle.com/kmldas/new-basline-np-log2-ensemble-top-10\">New Basline np.log2 Ensemble (top 10%)</a>\n\n<a href=\"https://www.kaggle.com/paklau9/minmax-highest-public-lb-9619\">MinMax highest public LB=.9619</a>\n\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","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Approach 1","execution_count":null},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"data1 = pd.read_csv('../input/siimmy-submissions/submission (3).csv')\n\ndata2 = pd.read_csv('../input/stacking-ensemble-on-my-submissions/submission_mean.csv')\n\ndata3 = pd.read_csv('../input/stacking-ensemble-on-my-submissions/submission_median.csv')\n\ndata4 = pd.read_csv('../input/eda-modelling-of-the-external-data-inc-ensemble/external_meta_ensembled.csv')\n\ndata5 = pd.read_csv('../input/new-basline-np-log2-ensemble-top-10/submission.csv')\n\ndata6 = pd.read_csv('../input/minmax-highest-public-lb-9619/submission.csv')\n\nsubmission = data1.copy()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"ncol = data1.shape[1]\n\ndata2['target'] = data4.iloc[:, 1:ncol].mean(axis=1)\ndata3['target'] = data4.iloc[:, 1:ncol].median(axis=1)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission['target'] =  1/6 * data1['target'] + 1/6 * data2['target'] + 1/6 * data3['target'] + 1/6 *data4['target']+ 1/6 *data5['target']+ 1/6 *data6['target']","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission.to_csv('submission.csv', index=False, float_format='%.6f')","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Approach 2","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"s1= pd.read_csv('../input/siimmy-submissions/submission (6).csv')\ns2= pd.read_csv('../input/siimmy-submissions/submission (7).csv')\ns3= pd.read_csv('../input/siimmy-submissions/b3submission.csv')\n\nsubmission= s1.copy()\n\nsubmission['target'] = 1/3 * s1['target'] + 1/3 * s2['target'] + 1/3 * s3['target'] \n\nsubmission.to_csv('b3submission.csv', index=False, float_format='%.6f')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","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}