{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import pandas as pd\nfrom IPython.display import display\nfrom sklearn.metrics import matthews_corrcoef, confusion_matrix\nimport warnings\nwarnings.simplefilter(action='ignore')\n\npd.set_option(\"display.max_rows\", 200)\npd.options.display.float_format = '{:,.2f}'.format","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"4c92b517618164d23bc8c19faa96c3788b645bff"},"cell_type":"code","source":"meta_train = pd.read_csv(\"../input/metadata_train.csv\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"088897c123a10838361d977c9fc7bce3f81e0b82"},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"cb128a206604cdb84b0d600b98ab934cd3c62d0c","scrolled":false},"cell_type":"code","source":"n_data = meta_train.shape[0]\nn_pos = meta_train.target.sum()\nprint(f\"n_data: {n_data}, n_pos: {n_pos}\")\n\ngt = [0]*(n_data-n_pos) + [1]*n_pos\npred = [0] * len(gt)  # initial prediction is all 0\n\nresult = []\nTN, FP, FN, TP = confusion_matrix(gt, pred).flatten().tolist()\nresult.append([TN, FP, FN, TP , matthews_corrcoef(gt, pred)])\n#result.append([0, matthews_corrcoef(gt, pred)])\nfor i in range(1, n_pos+1):\n    pred[-i] = 1\n    TN, FP, FN, TP = confusion_matrix(gt, pred).flatten().tolist()\n    result.append([TN, FP, FN, TP , matthews_corrcoef(gt, pred)])\n    \nprint(\"incleasing True Positive\")\ndf_score = pd.DataFrame(result, columns=[\"TrueNegative\", \"FalsePositive\", \"FalseNegative\", \"TruePositive\", \"MCC\"])\ndisplay(df_score.iloc[::5,:])\n\n\nresult = []\nTN, FP, FN, TP = confusion_matrix(gt, pred).flatten().tolist()\nresult.append([TN, FP, FN, TP , matthews_corrcoef(gt, pred)])\n#result.append([0, matthews_corrcoef(gt, pred)])\n\nfor i in range(1, n_pos*3+1):\n    pred[i] = 1\n    TN, FP, FN, TP = confusion_matrix(gt, pred).flatten().tolist()\n    result.append([TN, FP, FN, TP , matthews_corrcoef(gt, pred)])\n    \nprint()\nprint(\"incleasing False Positive\")\ndf_score = pd.DataFrame(result, columns=[\"TrueNegative\", \"FalsePositive\", \"FalseNegative\", \"TruePositive\", \"MCC\"])\ndisplay(df_score.iloc[::10,:])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"02307d6815e38dd64feba68d2b6fd8b0966736db"},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"39d5081dee1d9e3eed221f24b4858ccff5c76da6"},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}