{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.7.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"!nvidia-smi","metadata":{"execution":{"iopub.status.busy":"2023-03-26T08:07:18.582601Z","iopub.execute_input":"2023-03-26T08:07:18.583108Z","iopub.status.idle":"2023-03-26T08:07:19.925028Z","shell.execute_reply.started":"2023-03-26T08:07:18.583003Z","shell.execute_reply":"2023-03-26T08:07:19.923759Z"},"trusted":true},"execution_count":1,"outputs":[{"name":"stdout","text":"Sun Mar 26 08:07:19 2023       \n+-----------------------------------------------------------------------------+\n| NVIDIA-SMI 470.82.01    Driver Version: 470.82.01    CUDA Version: 11.4     |\n|-------------------------------+----------------------+----------------------+\n| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n|                               |                      |               MIG M. |\n|===============================+======================+======================|\n|   0  Tesla P100-PCIE...  Off  | 00000000:00:04.0 Off |                    0 |\n| N/A   37C    P0    26W / 250W |      0MiB / 16280MiB |      0%      Default |\n|                               |                      |                  N/A |\n+-------------------------------+----------------------+----------------------+\n                                                                               \n+-----------------------------------------------------------------------------+\n| Processes:                                                                  |\n|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n|        ID   ID                                                   Usage      |\n|=============================================================================|\n|  No running processes found                                                 |\n+-----------------------------------------------------------------------------+\n","output_type":"stream"}]},{"cell_type":"code","source":"!pip install timm","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:51:17.378749Z","iopub.execute_input":"2023-03-14T01:51:17.379121Z","iopub.status.idle":"2023-03-14T01:51:26.466801Z","shell.execute_reply.started":"2023-03-14T01:51:17.379089Z","shell.execute_reply":"2023-03-14T01:51:26.465679Z"},"trusted":true},"execution_count":3,"outputs":[{"name":"stdout","text":"Requirement already satisfied: timm in /opt/conda/lib/python3.7/site-packages (0.6.12)\nRequirement already satisfied: torch>=1.7 in /opt/conda/lib/python3.7/site-packages (from timm) (1.11.0)\nRequirement already satisfied: huggingface-hub in /opt/conda/lib/python3.7/site-packages (from timm) (0.10.1)\nRequirement already satisfied: pyyaml in /opt/conda/lib/python3.7/site-packages (from timm) (6.0)\nRequirement already satisfied: torchvision in /opt/conda/lib/python3.7/site-packages (from timm) (0.12.0)\nRequirement already satisfied: typing-extensions in /opt/conda/lib/python3.7/site-packages (from torch>=1.7->timm) (4.1.1)\nRequirement already satisfied: tqdm in /opt/conda/lib/python3.7/site-packages (from huggingface-hub->timm) (4.64.0)\nRequirement already satisfied: importlib-metadata in /opt/conda/lib/python3.7/site-packages (from huggingface-hub->timm) (4.13.0)\nRequirement already satisfied: packaging>=20.9 in /opt/conda/lib/python3.7/site-packages (from huggingface-hub->timm) (21.3)\nRequirement already satisfied: requests in /opt/conda/lib/python3.7/site-packages (from huggingface-hub->timm) (2.28.1)\nRequirement already satisfied: filelock in /opt/conda/lib/python3.7/site-packages (from huggingface-hub->timm) (3.7.1)\nRequirement already satisfied: numpy in /opt/conda/lib/python3.7/site-packages (from torchvision->timm) (1.21.6)\nRequirement already satisfied: pillow!=8.3.*,>=5.3.0 in /opt/conda/lib/python3.7/site-packages (from torchvision->timm) (9.1.1)\nRequirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /opt/conda/lib/python3.7/site-packages (from packaging>=20.9->huggingface-hub->timm) (3.0.9)\nRequirement already satisfied: zipp>=0.5 in /opt/conda/lib/python3.7/site-packages (from importlib-metadata->huggingface-hub->timm) (3.8.0)\nRequirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.7/site-packages (from requests->huggingface-hub->timm) (2022.9.24)\nRequirement already satisfied: charset-normalizer<3,>=2 in /opt/conda/lib/python3.7/site-packages (from requests->huggingface-hub->timm) (2.1.0)\nRequirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.7/site-packages (from requests->huggingface-hub->timm) (1.26.12)\nRequirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.7/site-packages (from requests->huggingface-hub->timm) (3.3)\n\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n\u001b[0m","output_type":"stream"}]},{"cell_type":"code","source":"import timm\navail_pretrained_models = timm.list_models(pretrained=True)\navail_pretrained_models","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:51:49.637423Z","iopub.execute_input":"2023-03-14T01:51:49.637788Z","iopub.status.idle":"2023-03-14T01:51:52.473252Z","shell.execute_reply.started":"2023-03-14T01:51:49.637756Z","shell.execute_reply":"2023-03-14T01:51:52.472196Z"},"trusted":true},"execution_count":5,"outputs":[{"execution_count":5,"output_type":"execute_result","data":{"text/plain":"['adv_inception_v3',\n 'bat_resnext26ts',\n 'beit_base_patch16_224',\n 'beit_base_patch16_224_in22k',\n 'beit_base_patch16_384',\n 'beit_large_patch16_224',\n 'beit_large_patch16_224_in22k',\n 'beit_large_patch16_384',\n 'beit_large_patch16_512',\n 'beitv2_base_patch16_224',\n 'beitv2_base_patch16_224_in22k',\n 'beitv2_large_patch16_224',\n 'beitv2_large_patch16_224_in22k',\n 'botnet26t_256',\n 'cait_m36_384',\n 'cait_m48_448',\n 'cait_s24_224',\n 'cait_s24_384',\n 'cait_s36_384',\n 'cait_xs24_384',\n 'cait_xxs24_224',\n 'cait_xxs24_384',\n 'cait_xxs36_224',\n 'cait_xxs36_384',\n 'coat_lite_mini',\n 'coat_lite_small',\n 'coat_lite_tiny',\n 'coat_mini',\n 'coat_tiny',\n 'coatnet_0_rw_224',\n 'coatnet_1_rw_224',\n 'coatnet_bn_0_rw_224',\n 'coatnet_nano_rw_224',\n 'coatnet_rmlp_1_rw_224',\n 'coatnet_rmlp_2_rw_224',\n 'coatnet_rmlp_nano_rw_224',\n 'coatnext_nano_rw_224',\n 'convit_base',\n 'convit_small',\n 'convit_tiny',\n 'convmixer_768_32',\n 'convmixer_1024_20_ks9_p14',\n 'convmixer_1536_20',\n 'convnext_atto',\n 'convnext_atto_ols',\n 'convnext_base',\n 'convnext_base_384_in22ft1k',\n 'convnext_base_in22ft1k',\n 'convnext_base_in22k',\n 'convnext_femto',\n 'convnext_femto_ols',\n 'convnext_large',\n 'convnext_large_384_in22ft1k',\n 'convnext_large_in22ft1k',\n 'convnext_large_in22k',\n 'convnext_nano',\n 'convnext_nano_ols',\n 'convnext_pico',\n 'convnext_pico_ols',\n 'convnext_small',\n 'convnext_small_384_in22ft1k',\n 'convnext_small_in22ft1k',\n 'convnext_small_in22k',\n 'convnext_tiny',\n 'convnext_tiny_384_in22ft1k',\n 'convnext_tiny_hnf',\n 'convnext_tiny_in22ft1k',\n 'convnext_tiny_in22k',\n 'convnext_xlarge_384_in22ft1k',\n 'convnext_xlarge_in22ft1k',\n 'convnext_xlarge_in22k',\n 'crossvit_9_240',\n 'crossvit_9_dagger_240',\n 'crossvit_15_240',\n 'crossvit_15_dagger_240',\n 'crossvit_15_dagger_408',\n 'crossvit_18_240',\n 'crossvit_18_dagger_240',\n 'crossvit_18_dagger_408',\n 'crossvit_base_240',\n 'crossvit_small_240',\n 'crossvit_tiny_240',\n 'cs3darknet_focus_l',\n 'cs3darknet_focus_m',\n 'cs3darknet_l',\n 'cs3darknet_m',\n 'cs3darknet_x',\n 'cs3edgenet_x',\n 'cs3se_edgenet_x',\n 'cs3sedarknet_l',\n 'cs3sedarknet_x',\n 'cspdarknet53',\n 'cspresnet50',\n 'cspresnext50',\n 'darknet53',\n 'darknetaa53',\n 'deit3_base_patch16_224',\n 'deit3_base_patch16_224_in21ft1k',\n 'deit3_base_patch16_384',\n 'deit3_base_patch16_384_in21ft1k',\n 'deit3_huge_patch14_224',\n 'deit3_huge_patch14_224_in21ft1k',\n 'deit3_large_patch16_224',\n 'deit3_large_patch16_224_in21ft1k',\n 'deit3_large_patch16_384',\n 'deit3_large_patch16_384_in21ft1k',\n 'deit3_medium_patch16_224',\n 'deit3_medium_patch16_224_in21ft1k',\n 'deit3_small_patch16_224',\n 'deit3_small_patch16_224_in21ft1k',\n 'deit3_small_patch16_384',\n 'deit3_small_patch16_384_in21ft1k',\n 'deit_base_distilled_patch16_224',\n 'deit_base_distilled_patch16_384',\n 'deit_base_patch16_224',\n 'deit_base_patch16_384',\n 'deit_small_distilled_patch16_224',\n 'deit_small_patch16_224',\n 'deit_tiny_distilled_patch16_224',\n 'deit_tiny_patch16_224',\n 'densenet121',\n 'densenet161',\n 'densenet169',\n 'densenet201',\n 'densenetblur121d',\n 'dla34',\n 'dla46_c',\n 'dla46x_c',\n 'dla60',\n 'dla60_res2net',\n 'dla60_res2next',\n 'dla60x',\n 'dla60x_c',\n 'dla102',\n 'dla102x',\n 'dla102x2',\n 'dla169',\n 'dm_nfnet_f0',\n 'dm_nfnet_f1',\n 'dm_nfnet_f2',\n 'dm_nfnet_f3',\n 'dm_nfnet_f4',\n 'dm_nfnet_f5',\n 'dm_nfnet_f6',\n 'dpn68',\n 'dpn68b',\n 'dpn92',\n 'dpn98',\n 'dpn107',\n 'dpn131',\n 'eca_botnext26ts_256',\n 'eca_halonext26ts',\n 'eca_nfnet_l0',\n 'eca_nfnet_l1',\n 'eca_nfnet_l2',\n 'eca_resnet33ts',\n 'eca_resnext26ts',\n 'ecaresnet26t',\n 'ecaresnet50d',\n 'ecaresnet50d_pruned',\n 'ecaresnet50t',\n 'ecaresnet101d',\n 'ecaresnet101d_pruned',\n 'ecaresnet269d',\n 'ecaresnetlight',\n 'edgenext_base',\n 'edgenext_small',\n 'edgenext_small_rw',\n 'edgenext_x_small',\n 'edgenext_xx_small',\n 'efficientformer_l1',\n 'efficientformer_l3',\n 'efficientformer_l7',\n 'efficientnet_b0',\n 'efficientnet_b1',\n 'efficientnet_b1_pruned',\n 'efficientnet_b2',\n 'efficientnet_b2_pruned',\n 'efficientnet_b3',\n 'efficientnet_b3_pruned',\n 'efficientnet_b4',\n 'efficientnet_el',\n 'efficientnet_el_pruned',\n 'efficientnet_em',\n 'efficientnet_es',\n 'efficientnet_es_pruned',\n 'efficientnet_lite0',\n 'efficientnetv2_rw_m',\n 'efficientnetv2_rw_s',\n 'efficientnetv2_rw_t',\n 'ens_adv_inception_resnet_v2',\n 'ese_vovnet19b_dw',\n 'ese_vovnet39b',\n 'fbnetc_100',\n 'fbnetv3_b',\n 'fbnetv3_d',\n 'fbnetv3_g',\n 'gc_efficientnetv2_rw_t',\n 'gcresnet33ts',\n 'gcresnet50t',\n 'gcresnext26ts',\n 'gcresnext50ts',\n 'gcvit_base',\n 'gcvit_small',\n 'gcvit_tiny',\n 'gcvit_xtiny',\n 'gcvit_xxtiny',\n 'gernet_l',\n 'gernet_m',\n 'gernet_s',\n 'ghostnet_100',\n 'gluon_inception_v3',\n 'gluon_resnet18_v1b',\n 'gluon_resnet34_v1b',\n 'gluon_resnet50_v1b',\n 'gluon_resnet50_v1c',\n 'gluon_resnet50_v1d',\n 'gluon_resnet50_v1s',\n 'gluon_resnet101_v1b',\n 'gluon_resnet101_v1c',\n 'gluon_resnet101_v1d',\n 'gluon_resnet101_v1s',\n 'gluon_resnet152_v1b',\n 'gluon_resnet152_v1c',\n 'gluon_resnet152_v1d',\n 'gluon_resnet152_v1s',\n 'gluon_resnext50_32x4d',\n 'gluon_resnext101_32x4d',\n 'gluon_resnext101_64x4d',\n 'gluon_senet154',\n 'gluon_seresnext50_32x4d',\n 'gluon_seresnext101_32x4d',\n 'gluon_seresnext101_64x4d',\n 'gluon_xception65',\n 'gmixer_24_224',\n 'gmlp_s16_224',\n 'halo2botnet50ts_256',\n 'halonet26t',\n 'halonet50ts',\n 'haloregnetz_b',\n 'hardcorenas_a',\n 'hardcorenas_b',\n 'hardcorenas_c',\n 'hardcorenas_d',\n 'hardcorenas_e',\n 'hardcorenas_f',\n 'hrnet_w18',\n 'hrnet_w18_small',\n 'hrnet_w18_small_v2',\n 'hrnet_w30',\n 'hrnet_w32',\n 'hrnet_w40',\n 'hrnet_w44',\n 'hrnet_w48',\n 'hrnet_w64',\n 'ig_resnext101_32x8d',\n 'ig_resnext101_32x16d',\n 'ig_resnext101_32x32d',\n 'ig_resnext101_32x48d',\n 'inception_resnet_v2',\n 'inception_v3',\n 'inception_v4',\n 'jx_nest_base',\n 'jx_nest_small',\n 'jx_nest_tiny',\n 'lambda_resnet26rpt_256',\n 'lambda_resnet26t',\n 'lambda_resnet50ts',\n 'lamhalobotnet50ts_256',\n 'lcnet_050',\n 'lcnet_075',\n 'lcnet_100',\n 'legacy_senet154',\n 'legacy_seresnet18',\n 'legacy_seresnet34',\n 'legacy_seresnet50',\n 'legacy_seresnet101',\n 'legacy_seresnet152',\n 'legacy_seresnext26_32x4d',\n 'legacy_seresnext50_32x4d',\n 'legacy_seresnext101_32x4d',\n 'levit_128',\n 'levit_128s',\n 'levit_192',\n 'levit_256',\n 'levit_384',\n 'maxvit_nano_rw_256',\n 'maxvit_rmlp_nano_rw_256',\n 'maxvit_rmlp_pico_rw_256',\n 'maxvit_rmlp_small_rw_224',\n 'maxvit_rmlp_tiny_rw_256',\n 'maxvit_tiny_rw_224',\n 'maxxvit_rmlp_nano_rw_256',\n 'maxxvit_rmlp_small_rw_256',\n 'mixer_b16_224',\n 'mixer_b16_224_in21k',\n 'mixer_b16_224_miil',\n 'mixer_b16_224_miil_in21k',\n 'mixer_l16_224',\n 'mixer_l16_224_in21k',\n 'mixnet_l',\n 'mixnet_m',\n 'mixnet_s',\n 'mixnet_xl',\n 'mnasnet_100',\n 'mnasnet_small',\n 'mobilenetv2_050',\n 'mobilenetv2_100',\n 'mobilenetv2_110d',\n 'mobilenetv2_120d',\n 'mobilenetv2_140',\n 'mobilenetv3_large_100',\n 'mobilenetv3_large_100_miil',\n 'mobilenetv3_large_100_miil_in21k',\n 'mobilenetv3_rw',\n 'mobilenetv3_small_050',\n 'mobilenetv3_small_075',\n 'mobilenetv3_small_100',\n 'mobilevit_s',\n 'mobilevit_xs',\n 'mobilevit_xxs',\n 'mobilevitv2_050',\n 'mobilevitv2_075',\n 'mobilevitv2_100',\n 'mobilevitv2_125',\n 'mobilevitv2_150',\n 'mobilevitv2_150_384_in22ft1k',\n 'mobilevitv2_150_in22ft1k',\n 'mobilevitv2_175',\n 'mobilevitv2_175_384_in22ft1k',\n 'mobilevitv2_175_in22ft1k',\n 'mobilevitv2_200',\n 'mobilevitv2_200_384_in22ft1k',\n 'mobilevitv2_200_in22ft1k',\n 'mvitv2_base',\n 'mvitv2_large',\n 'mvitv2_small',\n 'mvitv2_tiny',\n 'nasnetalarge',\n 'nf_regnet_b1',\n 'nf_resnet50',\n 'nfnet_l0',\n 'pit_b_224',\n 'pit_b_distilled_224',\n 'pit_s_224',\n 'pit_s_distilled_224',\n 'pit_ti_224',\n 'pit_ti_distilled_224',\n 'pit_xs_224',\n 'pit_xs_distilled_224',\n 'pnasnet5large',\n 'poolformer_m36',\n 'poolformer_m48',\n 'poolformer_s12',\n 'poolformer_s24',\n 'poolformer_s36',\n 'pvt_v2_b0',\n 'pvt_v2_b1',\n 'pvt_v2_b2',\n 'pvt_v2_b2_li',\n 'pvt_v2_b3',\n 'pvt_v2_b4',\n 'pvt_v2_b5',\n 'regnetv_040',\n 'regnetv_064',\n 'regnetx_002',\n 'regnetx_004',\n 'regnetx_006',\n 'regnetx_008',\n 'regnetx_016',\n 'regnetx_032',\n 'regnetx_040',\n 'regnetx_064',\n 'regnetx_080',\n 'regnetx_120',\n 'regnetx_160',\n 'regnetx_320',\n 'regnety_002',\n 'regnety_004',\n 'regnety_006',\n 'regnety_008',\n 'regnety_016',\n 'regnety_032',\n 'regnety_040',\n 'regnety_064',\n 'regnety_080',\n 'regnety_120',\n 'regnety_160',\n 'regnety_320',\n 'regnetz_040',\n 'regnetz_040h',\n 'regnetz_b16',\n 'regnetz_c16',\n 'regnetz_c16_evos',\n 'regnetz_d8',\n 'regnetz_d8_evos',\n 'regnetz_d32',\n 'regnetz_e8',\n 'repvgg_a2',\n 'repvgg_b0',\n 'repvgg_b1',\n 'repvgg_b1g4',\n 'repvgg_b2',\n 'repvgg_b2g4',\n 'repvgg_b3',\n 'repvgg_b3g4',\n 'res2net50_14w_8s',\n 'res2net50_26w_4s',\n 'res2net50_26w_6s',\n 'res2net50_26w_8s',\n 'res2net50_48w_2s',\n 'res2net101_26w_4s',\n 'res2next50',\n 'resmlp_12_224',\n 'resmlp_12_224_dino',\n 'resmlp_12_distilled_224',\n 'resmlp_24_224',\n 'resmlp_24_224_dino',\n 'resmlp_24_distilled_224',\n 'resmlp_36_224',\n 'resmlp_36_distilled_224',\n 'resmlp_big_24_224',\n 'resmlp_big_24_224_in22ft1k',\n 'resmlp_big_24_distilled_224',\n 'resnest14d',\n 'resnest26d',\n 'resnest50d',\n 'resnest50d_1s4x24d',\n 'resnest50d_4s2x40d',\n 'resnest101e',\n 'resnest200e',\n 'resnest269e',\n 'resnet10t',\n 'resnet14t',\n 'resnet18',\n 'resnet18d',\n 'resnet26',\n 'resnet26d',\n 'resnet26t',\n 'resnet32ts',\n 'resnet33ts',\n 'resnet34',\n 'resnet34d',\n 'resnet50',\n 'resnet50_gn',\n 'resnet50d',\n 'resnet51q',\n 'resnet61q',\n 'resnet101',\n 'resnet101d',\n 'resnet152',\n 'resnet152d',\n 'resnet200d',\n 'resnetaa50',\n 'resnetblur50',\n 'resnetrs50',\n 'resnetrs101',\n 'resnetrs152',\n 'resnetrs200',\n 'resnetrs270',\n 'resnetrs350',\n 'resnetrs420',\n 'resnetv2_50',\n 'resnetv2_50d_evos',\n 'resnetv2_50d_gn',\n 'resnetv2_50x1_bit_distilled',\n 'resnetv2_50x1_bitm',\n 'resnetv2_50x1_bitm_in21k',\n 'resnetv2_50x3_bitm',\n 'resnetv2_50x3_bitm_in21k',\n 'resnetv2_101',\n 'resnetv2_101x1_bitm',\n 'resnetv2_101x1_bitm_in21k',\n 'resnetv2_101x3_bitm',\n 'resnetv2_101x3_bitm_in21k',\n 'resnetv2_152x2_bit_teacher',\n 'resnetv2_152x2_bit_teacher_384',\n 'resnetv2_152x2_bitm',\n 'resnetv2_152x2_bitm_in21k',\n 'resnetv2_152x4_bitm',\n 'resnetv2_152x4_bitm_in21k',\n 'resnext26ts',\n 'resnext50_32x4d',\n 'resnext50d_32x4d',\n 'resnext101_32x8d',\n 'resnext101_64x4d',\n 'rexnet_100',\n 'rexnet_130',\n 'rexnet_150',\n 'rexnet_200',\n 'sebotnet33ts_256',\n 'sehalonet33ts',\n 'selecsls42b',\n 'selecsls60',\n 'selecsls60b',\n 'semnasnet_075',\n 'semnasnet_100',\n 'sequencer2d_l',\n 'sequencer2d_m',\n 'sequencer2d_s',\n 'seresnet33ts',\n 'seresnet50',\n 'seresnet152d',\n 'seresnext26d_32x4d',\n 'seresnext26t_32x4d',\n 'seresnext26ts',\n 'seresnext50_32x4d',\n 'seresnext101_32x8d',\n 'seresnext101d_32x8d',\n 'seresnextaa101d_32x8d',\n 'skresnet18',\n 'skresnet34',\n 'skresnext50_32x4d',\n 'spnasnet_100',\n 'ssl_resnet18',\n 'ssl_resnet50',\n 'ssl_resnext50_32x4d',\n 'ssl_resnext101_32x4d',\n 'ssl_resnext101_32x8d',\n 'ssl_resnext101_32x16d',\n 'swin_base_patch4_window7_224',\n 'swin_base_patch4_window7_224_in22k',\n 'swin_base_patch4_window12_384',\n 'swin_base_patch4_window12_384_in22k',\n 'swin_large_patch4_window7_224',\n 'swin_large_patch4_window7_224_in22k',\n 'swin_large_patch4_window12_384',\n 'swin_large_patch4_window12_384_in22k',\n 'swin_s3_base_224',\n 'swin_s3_small_224',\n 'swin_s3_tiny_224',\n 'swin_small_patch4_window7_224',\n 'swin_tiny_patch4_window7_224',\n 'swinv2_base_window8_256',\n 'swinv2_base_window12_192_22k',\n 'swinv2_base_window12to16_192to256_22kft1k',\n 'swinv2_base_window12to24_192to384_22kft1k',\n 'swinv2_base_window16_256',\n 'swinv2_cr_small_224',\n 'swinv2_cr_small_ns_224',\n 'swinv2_cr_tiny_ns_224',\n 'swinv2_large_window12_192_22k',\n 'swinv2_large_window12to16_192to256_22kft1k',\n 'swinv2_large_window12to24_192to384_22kft1k',\n 'swinv2_small_window8_256',\n 'swinv2_small_window16_256',\n 'swinv2_tiny_window8_256',\n 'swinv2_tiny_window16_256',\n 'swsl_resnet18',\n 'swsl_resnet50',\n 'swsl_resnext50_32x4d',\n 'swsl_resnext101_32x4d',\n 'swsl_resnext101_32x8d',\n 'swsl_resnext101_32x16d',\n 'tf_efficientnet_b0',\n 'tf_efficientnet_b0_ap',\n 'tf_efficientnet_b0_ns',\n 'tf_efficientnet_b1',\n 'tf_efficientnet_b1_ap',\n 'tf_efficientnet_b1_ns',\n 'tf_efficientnet_b2',\n 'tf_efficientnet_b2_ap',\n 'tf_efficientnet_b2_ns',\n 'tf_efficientnet_b3',\n 'tf_efficientnet_b3_ap',\n 'tf_efficientnet_b3_ns',\n 'tf_efficientnet_b4',\n 'tf_efficientnet_b4_ap',\n 'tf_efficientnet_b4_ns',\n 'tf_efficientnet_b5',\n 'tf_efficientnet_b5_ap',\n 'tf_efficientnet_b5_ns',\n 'tf_efficientnet_b6',\n 'tf_efficientnet_b6_ap',\n 'tf_efficientnet_b6_ns',\n 'tf_efficientnet_b7',\n 'tf_efficientnet_b7_ap',\n 'tf_efficientnet_b7_ns',\n 'tf_efficientnet_b8',\n 'tf_efficientnet_b8_ap',\n 'tf_efficientnet_cc_b0_4e',\n 'tf_efficientnet_cc_b0_8e',\n 'tf_efficientnet_cc_b1_8e',\n 'tf_efficientnet_el',\n 'tf_efficientnet_em',\n 'tf_efficientnet_es',\n 'tf_efficientnet_l2_ns',\n 'tf_efficientnet_l2_ns_475',\n 'tf_efficientnet_lite0',\n 'tf_efficientnet_lite1',\n 'tf_efficientnet_lite2',\n 'tf_efficientnet_lite3',\n 'tf_efficientnet_lite4',\n 'tf_efficientnetv2_b0',\n 'tf_efficientnetv2_b1',\n 'tf_efficientnetv2_b2',\n 'tf_efficientnetv2_b3',\n 'tf_efficientnetv2_l',\n 'tf_efficientnetv2_l_in21ft1k',\n 'tf_efficientnetv2_l_in21k',\n 'tf_efficientnetv2_m',\n 'tf_efficientnetv2_m_in21ft1k',\n 'tf_efficientnetv2_m_in21k',\n 'tf_efficientnetv2_s',\n 'tf_efficientnetv2_s_in21ft1k',\n 'tf_efficientnetv2_s_in21k',\n 'tf_efficientnetv2_xl_in21ft1k',\n 'tf_efficientnetv2_xl_in21k',\n 'tf_inception_v3',\n 'tf_mixnet_l',\n 'tf_mixnet_m',\n 'tf_mixnet_s',\n 'tf_mobilenetv3_large_075',\n 'tf_mobilenetv3_large_100',\n 'tf_mobilenetv3_large_minimal_100',\n 'tf_mobilenetv3_small_075',\n 'tf_mobilenetv3_small_100',\n 'tf_mobilenetv3_small_minimal_100',\n 'tinynet_a',\n 'tinynet_b',\n 'tinynet_c',\n 'tinynet_d',\n 'tinynet_e',\n 'tnt_s_patch16_224',\n 'tresnet_l',\n 'tresnet_l_448',\n 'tresnet_m',\n 'tresnet_m_448',\n 'tresnet_m_miil_in21k',\n 'tresnet_v2_l',\n 'tresnet_xl',\n 'tresnet_xl_448',\n 'tv_densenet121',\n 'tv_resnet34',\n 'tv_resnet50',\n 'tv_resnet101',\n 'tv_resnet152',\n 'tv_resnext50_32x4d',\n 'twins_pcpvt_base',\n 'twins_pcpvt_large',\n 'twins_pcpvt_small',\n 'twins_svt_base',\n 'twins_svt_large',\n 'twins_svt_small',\n 'vgg11',\n 'vgg11_bn',\n 'vgg13',\n 'vgg13_bn',\n 'vgg16',\n 'vgg16_bn',\n 'vgg19',\n 'vgg19_bn',\n 'visformer_small',\n 'vit_base_patch8_224',\n 'vit_base_patch8_224_dino',\n 'vit_base_patch8_224_in21k',\n 'vit_base_patch16_224',\n 'vit_base_patch16_224_dino',\n 'vit_base_patch16_224_in21k',\n 'vit_base_patch16_224_miil',\n 'vit_base_patch16_224_miil_in21k',\n 'vit_base_patch16_224_sam',\n 'vit_base_patch16_384',\n 'vit_base_patch16_rpn_224',\n 'vit_base_patch32_224',\n 'vit_base_patch32_224_clip_laion2b',\n 'vit_base_patch32_224_in21k',\n 'vit_base_patch32_224_sam',\n 'vit_base_patch32_384',\n 'vit_base_r50_s16_224_in21k',\n 'vit_base_r50_s16_384',\n 'vit_giant_patch14_224_clip_laion2b',\n 'vit_huge_patch14_224_clip_laion2b',\n 'vit_huge_patch14_224_in21k',\n 'vit_large_patch14_224_clip_laion2b',\n 'vit_large_patch16_224',\n 'vit_large_patch16_224_in21k',\n 'vit_large_patch16_384',\n 'vit_large_patch32_224_in21k',\n 'vit_large_patch32_384',\n 'vit_large_r50_s32_224',\n 'vit_large_r50_s32_224_in21k',\n 'vit_large_r50_s32_384',\n 'vit_relpos_base_patch16_224',\n 'vit_relpos_base_patch16_clsgap_224',\n 'vit_relpos_base_patch32_plus_rpn_256',\n 'vit_relpos_medium_patch16_224',\n 'vit_relpos_medium_patch16_cls_224',\n 'vit_relpos_medium_patch16_rpn_224',\n 'vit_relpos_small_patch16_224',\n 'vit_small_patch8_224_dino',\n 'vit_small_patch16_224',\n 'vit_small_patch16_224_dino',\n 'vit_small_patch16_224_in21k',\n 'vit_small_patch16_384',\n 'vit_small_patch32_224',\n 'vit_small_patch32_224_in21k',\n 'vit_small_patch32_384',\n 'vit_small_r26_s32_224',\n 'vit_small_r26_s32_224_in21k',\n 'vit_small_r26_s32_384',\n 'vit_srelpos_medium_patch16_224',\n 'vit_srelpos_small_patch16_224',\n 'vit_tiny_patch16_224',\n 'vit_tiny_patch16_224_in21k',\n 'vit_tiny_patch16_384',\n 'vit_tiny_r_s16_p8_224',\n 'vit_tiny_r_s16_p8_224_in21k',\n 'vit_tiny_r_s16_p8_384',\n 'volo_d1_224',\n 'volo_d1_384',\n 'volo_d2_224',\n 'volo_d2_384',\n 'volo_d3_224',\n 'volo_d3_448',\n 'volo_d4_224',\n 'volo_d4_448',\n 'volo_d5_224',\n 'volo_d5_448',\n 'volo_d5_512',\n 'wide_resnet50_2',\n 'wide_resnet101_2',\n 'xception',\n 'xception41',\n 'xception41p',\n 'xception65',\n 'xception65p',\n 'xception71',\n 'xcit_large_24_p8_224',\n 'xcit_large_24_p8_224_dist',\n 'xcit_large_24_p8_384_dist',\n 'xcit_large_24_p16_224',\n 'xcit_large_24_p16_224_dist',\n 'xcit_large_24_p16_384_dist',\n 'xcit_medium_24_p8_224',\n 'xcit_medium_24_p8_224_dist',\n 'xcit_medium_24_p8_384_dist',\n 'xcit_medium_24_p16_224',\n 'xcit_medium_24_p16_224_dist',\n 'xcit_medium_24_p16_384_dist',\n 'xcit_nano_12_p8_224',\n 'xcit_nano_12_p8_224_dist',\n 'xcit_nano_12_p8_384_dist',\n 'xcit_nano_12_p16_224',\n 'xcit_nano_12_p16_224_dist',\n 'xcit_nano_12_p16_384_dist',\n 'xcit_small_12_p8_224',\n 'xcit_small_12_p8_224_dist',\n 'xcit_small_12_p8_384_dist',\n 'xcit_small_12_p16_224',\n 'xcit_small_12_p16_224_dist',\n 'xcit_small_12_p16_384_dist',\n 'xcit_small_24_p8_224',\n 'xcit_small_24_p8_224_dist',\n 'xcit_small_24_p8_384_dist',\n 'xcit_small_24_p16_224',\n 'xcit_small_24_p16_224_dist',\n 'xcit_small_24_p16_384_dist',\n 'xcit_tiny_12_p8_224',\n 'xcit_tiny_12_p8_224_dist',\n 'xcit_tiny_12_p8_384_dist',\n 'xcit_tiny_12_p16_224',\n 'xcit_tiny_12_p16_224_dist',\n 'xcit_tiny_12_p16_384_dist',\n 'xcit_tiny_24_p8_224',\n 'xcit_tiny_24_p8_224_dist',\n 'xcit_tiny_24_p8_384_dist',\n 'xcit_tiny_24_p16_224',\n 'xcit_tiny_24_p16_224_dist',\n 'xcit_tiny_24_p16_384_dist']"},"metadata":{}}]},{"cell_type":"markdown","source":"# Imports","metadata":{}},{"cell_type":"code","source":"import math\nimport time\nfrom pathlib import Path\n\nimport numpy as np\nimport pandas as pd\nfrom sklearn.model_selection import StratifiedGroupKFold\n\nimport cv2\nimport timm\n\nimport torch\nimport torch.nn as nn\nimport torch.optim as optim\nfrom torch.utils.data import Dataset, DataLoader\n\nimport albumentations\nfrom albumentations.pytorch import ToTensorV2","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:31.228224Z","iopub.execute_input":"2023-03-14T01:56:31.228724Z","iopub.status.idle":"2023-03-14T01:56:32.591506Z","shell.execute_reply.started":"2023-03-14T01:56:31.228678Z","shell.execute_reply":"2023-03-14T01:56:32.590484Z"},"trusted":true},"execution_count":9,"outputs":[]},{"cell_type":"code","source":"!ls ../input/rsna-breast-cancer-detection/","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:32.593492Z","iopub.execute_input":"2023-03-14T01:56:32.594198Z","iopub.status.idle":"2023-03-14T01:56:33.611987Z","shell.execute_reply.started":"2023-03-14T01:56:32.594145Z","shell.execute_reply":"2023-03-14T01:56:33.610779Z"},"trusted":true},"execution_count":10,"outputs":[{"name":"stdout","text":"sample_submission.csv  test.csv  test_images  train.csv  train_images\n","output_type":"stream"}]},{"cell_type":"markdown","source":"# Config","metadata":{}},{"cell_type":"code","source":"BASE_IMG_DIR = \"../input/rsna-breast-cancer-256-pngs/\"\n\nclass CFG:\n    \n    model_name = \"resnet50d\"\n    n_folds = 5\n    n_classes = 1\n    n_epochs = 5\n    train_batch_size = 64\n    valid_batch_size = 64\n    lr = 1e-4\n    wd = 1e-6\n    gradient_accumulation_steps = 1\n    max_grad_norm = 1000\n    print_every = 100","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:33.614883Z","iopub.execute_input":"2023-03-14T01:56:33.615306Z","iopub.status.idle":"2023-03-14T01:56:33.621259Z","shell.execute_reply.started":"2023-03-14T01:56:33.615262Z","shell.execute_reply":"2023-03-14T01:56:33.620236Z"},"trusted":true},"execution_count":11,"outputs":[]},{"cell_type":"markdown","source":"# CV Split","metadata":{"execution":{"iopub.status.busy":"2022-11-30T16:22:03.991014Z","iopub.execute_input":"2022-11-30T16:22:03.991406Z","iopub.status.idle":"2022-11-30T16:22:04.059074Z","shell.execute_reply.started":"2022-11-30T16:22:03.991372Z","shell.execute_reply":"2022-11-30T16:22:04.05811Z"}}},{"cell_type":"code","source":"df_all = pd.read_csv(\"../input/rsna-breast-cancer-detection/train.csv\")\ndf_all[\"fold\"] = -1\n\ngkfold = StratifiedGroupKFold(n_splits=CFG.n_folds)\nfor fold_idx, (train_idx, val_idx) in enumerate(gkfold.split(df_all, y=df_all.cancer, groups=df_all.patient_id)):\n    df_all.loc[val_idx, \"fold\"] = fold_idx","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:33.62423Z","iopub.execute_input":"2023-03-14T01:56:33.624954Z","iopub.status.idle":"2023-03-14T01:56:37.821573Z","shell.execute_reply.started":"2023-03-14T01:56:33.624908Z","shell.execute_reply":"2023-03-14T01:56:37.820567Z"},"trusted":true},"execution_count":12,"outputs":[]},{"cell_type":"code","source":"df_all.groupby(\"fold\")[\"cancer\"].value_counts().plot(kind=\"bar\")","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:37.822921Z","iopub.execute_input":"2023-03-14T01:56:37.823766Z","iopub.status.idle":"2023-03-14T01:56:38.084103Z","shell.execute_reply.started":"2023-03-14T01:56:37.823728Z","shell.execute_reply":"2023-03-14T01:56:38.083082Z"},"trusted":true},"execution_count":13,"outputs":[{"execution_count":13,"output_type":"execute_result","data":{"text/plain":"<AxesSubplot:xlabel='fold,cancer'>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 1 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAAYMAAAEXCAYAAABPkyhHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAAsTAAALEwEAmpwYAAAUxUlEQVR4nO3df7DddX3n8edLIsW2ys/biEnYpDW1jba1kAFc3B0FC0Hshu0q0lGJLt20I4ruuuNid2eiIju405HK7kqHKWkDyxQp7ZaMyGIKSGunIEFafsqSokgimGAA3aLU4Hv/OJ/LPV7vTbj33Hvul5znY+bOPefz/Z7v93W/Nyeve77f7/meVBWSpNH2ooUOIElaeJaBJMkykCRZBpIkLANJEpaBJAlYtNABZuuII46o5cuXL3QMSXrBuOOOOx6vqrGppr1gy2D58uVs3bp1oWNI0gtGkoenm+ZuIkmSZSBJsgwkSVgGkiQsA0kSloEkCctAkoRlIEniBfyms31Zft51Ay/j6xeeNgdJFp7bYoLbYoLbYoLbwlcGkiQsA0kSloEkCctAkoRlIEnCMpAkYRlIkrAMJElYBpIknkcZJNmYZGeSe/rGDkuyJcmD7fuhbTxJLk6yLcldSY7ue8y6Nv+DSdb1jR+T5O72mIuTZK5/SEnS3j2fVwZ/DKyZNHYecGNVrQRubPcBTgVWtq/1wCXQKw9gA3AccCywYbxA2jz/ru9xk9clSZpn+yyDqvorYPek4bXApnZ7E3B63/jl1XMrcEiSI4FTgC1VtbuqngC2AGvatJdV1a1VVcDlfcuSJA3JbI8ZLK6qR9vtx4DF7fYS4JG++ba3sb2Nb59iXJI0RAMfQG5/0dccZNmnJOuTbE2yddeuXcNYpSSNhNmWwbfaLh7a951tfAewrG++pW1sb+NLpxifUlVdWlWrq2r12NjYLKNLkiabbRlsBsbPCFoHXNs3flY7q+h44Km2O+kG4OQkh7YDxycDN7Rp30lyfDuL6Ky+ZUmShmSfH26T5E+ANwBHJNlO76ygC4Grk5wNPAyc0Wb/PPBmYBvwNPAegKraneR84PY238eravyg9HvpnbH0EuD69iVJGqJ9lkFV/eY0k06aYt4CzplmORuBjVOMbwVes68ckqT54zuQJUmWgSTJMpAkYRlIkrAMJElYBpIkLANJEpaBJAnLQJKEZSBJwjKQJGEZSJKwDCRJWAaSJCwDSRKWgSQJy0CShGUgScIykCRhGUiSsAwkSVgGkiQsA0kSloEkCctAkoRlIEnCMpAkYRlIkrAMJElYBpIkBiyDJP8+yb1J7knyJ0kOSrIiyW1JtiX5bJID27w/0e5va9OX9y3nI238gSSnDPgzSZJmaNZlkGQJcC6wuqpeAxwAnAl8Erioql4JPAGc3R5yNvBEG7+ozUeSVe1xrwbWAJ9JcsBsc0mSZm7Q3USLgJckWQT8JPAocCJwTZu+CTi93V7b7tOmn5Qkbfyqqnqmqr4GbAOOHTCXJGkGZl0GVbUD+D3gG/RK4CngDuDJqtrTZtsOLGm3lwCPtMfuafMf3j8+xWN+RJL1SbYm2bpr167ZRpckTTLIbqJD6f1VvwJ4BfBT9HbzzJuqurSqVlfV6rGxsflclSSNlEF2E70J+FpV7aqqHwB/DpwAHNJ2GwEsBXa02zuAZQBt+sHAt/vHp3iMJGkIBimDbwDHJ/nJtu//JOA+4GbgrW2edcC17fbmdp82/aaqqjZ+ZjvbaAWwEvjyALkkSTO0aN+zTK2qbktyDfAVYA9wJ3ApcB1wVZJPtLHL2kMuA65Isg3YTe8MIqrq3iRX0yuSPcA5VfXsbHNJkmZu1mUAUFUbgA2Thh9iirOBqur7wNumWc4FwAWDZJEkzZ7vQJYkWQaSJMtAkoRlIEnCMpAkYRlIkrAMJElYBpIkLANJEpaBJAnLQJKEZSBJwjKQJGEZSJKwDCRJWAaSJCwDSRKWgSQJy0CShGUgScIykCRhGUiSsAwkSVgGkiQsA0kSloEkCctAkoRlIEnCMpAkYRlIkhiwDJIckuSaJF9Ncn+S1yU5LMmWJA+274e2eZPk4iTbktyV5Oi+5axr8z+YZN2gP5QkaWYGfWXwaeD/VNUvAL8C3A+cB9xYVSuBG9t9gFOBle1rPXAJQJLDgA3AccCxwIbxApEkDcesyyDJwcC/BC4DqKp/qqongbXApjbbJuD0dnstcHn13AockuRI4BRgS1XtrqongC3AmtnmkiTN3CCvDFYAu4A/SnJnkj9M8lPA4qp6tM3zGLC43V4CPNL3+O1tbLrxH5NkfZKtSbbu2rVrgOiSpH6DlMEi4Gjgkqr6VeAfmdglBEBVFVADrONHVNWlVbW6qlaPjY3N1WIlaeQNUgbbge1VdVu7fw29cvhW2/1D+76zTd8BLOt7/NI2Nt24JGlIZl0GVfUY8EiSV7Whk4D7gM3A+BlB64Br2+3NwFntrKLjgafa7qQbgJOTHNoOHJ/cxiRJQ7JowMe/H7gyyYHAQ8B76BXM1UnOBh4Gzmjzfh54M7ANeLrNS1XtTnI+cHub7+NVtXvAXJKkGRioDKrq74DVU0w6aYp5CzhnmuVsBDYOkkWSNHu+A1mSZBlIkiwDSRKWgSQJy0CShGUgScIykCRhGUiSsAwkSVgGkiQsA0kSloEkCctAkoRlIEnCMpAkYRlIkrAMJElYBpIkLANJEpaBJAnLQJKEZSBJwjKQJGEZSJKwDCRJWAaSJCwDSRKWgSQJy0CSxByUQZIDktyZ5HPt/ooktyXZluSzSQ5s4z/R7m9r05f3LeMjbfyBJKcMmkmSNDNz8crgA8D9ffc/CVxUVa8EngDObuNnA0+08YvafCRZBZwJvBpYA3wmyQFzkEuS9DwNVAZJlgKnAX/Y7gc4EbimzbIJOL3dXtvu06af1OZfC1xVVc9U1deAbcCxg+SSJM3MoK8Mfh/4MPDDdv9w4Mmq2tPubweWtNtLgEcA2vSn2vzPjU/xGEnSEMy6DJK8BdhZVXfMYZ59rXN9kq1Jtu7atWtYq5Wk/d4grwxOAP5Vkq8DV9HbPfRp4JAki9o8S4Ed7fYOYBlAm34w8O3+8Ske8yOq6tKqWl1Vq8fGxgaILknqN+syqKqPVNXSqlpO7wDwTVX1DuBm4K1ttnXAte325nafNv2mqqo2fmY722gFsBL48mxzSZJmbtG+Z5mx/wRcleQTwJ3AZW38MuCKJNuA3fQKhKq6N8nVwH3AHuCcqnp2HnJJkqYxJ2VQVV8EvthuP8QUZwNV1feBt03z+AuAC+YiiyRp5nwHsiTJMpAkWQaSJCwDSRKWgSQJy0CShGUgScIykCRhGUiSsAwkSVgGkiQsA0kSloEkCctAkoRlIEnCMpAkYRlIkrAMJElYBpIkLANJEpaBJAnLQJKEZSBJwjKQJGEZSJKwDCRJWAaSJCwDSRKWgSQJy0CSxABlkGRZkpuT3Jfk3iQfaOOHJdmS5MH2/dA2niQXJ9mW5K4kR/cta12b/8Ek6wb/sSRJMzHIK4M9wIeqahVwPHBOklXAecCNVbUSuLHdBzgVWNm+1gOXQK88gA3AccCxwIbxApEkDcesy6CqHq2qr7Tb3wXuB5YAa4FNbbZNwOnt9lrg8uq5FTgkyZHAKcCWqtpdVU8AW4A1s80lSZq5OTlmkGQ58KvAbcDiqnq0TXoMWNxuLwEe6XvY9jY23fhU61mfZGuSrbt27ZqL6JIk5qAMkvw08GfAB6vqO/3TqqqAGnQdfcu7tKpWV9XqsbGxuVqsJI28gcogyYvpFcGVVfXnbfhbbfcP7fvONr4DWNb38KVtbLpxSdKQDHI2UYDLgPur6lN9kzYD42cErQOu7Rs/q51VdDzwVNuddANwcpJD24Hjk9uYJGlIFg3w2BOAdwF3J/m7Nva7wIXA1UnOBh4GzmjTPg+8GdgGPA28B6Cqdic5H7i9zffxqto9QC5J0gzNugyq6ktAppl80hTzF3DONMvaCGycbRZJ0mB8B7IkyTKQJFkGkiQsA0kSloEkCctAkoRlIEnCMpAkYRlIkrAMJElYBpIkLANJEpaBJAnLQJKEZSBJwjKQJGEZSJKwDCRJWAaSJCwDSRKWgSQJy0CShGUgScIykCRhGUiSsAwkSVgGkiQsA0kSloEkCVi00AH2Z8vPu27gZXz9wtPmIMnCc1tMcFtMcFtMWOht0ZlXBknWJHkgybYk5y10HkkaJZ0ogyQHAP8TOBVYBfxmklULm0qSRkcnygA4FthWVQ9V1T8BVwFrFziTJI2MVNVCZyDJW4E1VfVb7f67gOOq6n2T5lsPrG93XwU8MMBqjwAeH+Dxc6ULObqQAbqRowsZoBs5upABupGjCxlg8Bz/rKrGpprwgjqAXFWXApfOxbKSbK2q1XOxrBd6ji5k6EqOLmToSo4uZOhKji5kmO8cXdlNtANY1nd/aRuTJA1BV8rgdmBlkhVJDgTOBDYvcCZJGhmd2E1UVXuSvA+4ATgA2FhV987zaudkd9Mc6EKOLmSAbuToQgboRo4uZIBu5OhCBpjHHJ04gCxJWlhd2U0kSVpAloEkyTKQJHXkAPIwJDkIeAvwL4BXAN8D7gGuG8LB6vEMS+mdKfVjGYDrq+qHo5IjyeuAd7YMR07K8L+q6qn5ztByuC0mcvgcmcgxcttiJA4gJ/kYvV/sF4E7gJ3AQcDPA29stz9UVXfNY4Y/ApYAnwO2TpHhGOC8qvqr+crQlRxJrge+CVw7TYZfBz5VVfN6erHb4kdy+ByZyDGS22JUyuC0qpr2+rBJfgY4qqq2zmOG11TVPXuZfmDLsG2+MnQlR5Ijqmqvb6l/PvPMQQ63xcQ6fI5MrGckt8VIlIGmluQwgKravdBZFprbQqNuJA4gJzk4yYVJvppkd5JvJ7m/jR3SgXzXD3FdRyW5Ksku4Dbgy0l2trHlw8oxnSR3D3FdbouJdfkcmVjXSG6LUTmAfDVwE/CGqnoMIMnLgXVt2snzHSDJ0dNNAl473+vv81ng94F3VNWz8NznSbyN3qXDj5/vAEl+Y7pJwMvne/193BYTfI5MGMltMRK7iZI8UFWvmum0Oc7wLHALvV/mZMdX1UvmO0PL8WBVrZzptDnO8APgSmCqf3xvraqXzneGlsNtMZHD58hEjpHcFqPyyuDhJB8GNlXVtwCSLAbeDTwypAz3A79dVQ9OnpBkWBkA7kjyGWATEz/7Mnp/9dw5pAx3Ab831QGyJG8aUgZwW/TzOTJhJLfFSBwzAN4OHA7c0vYB7qZ32thhwBlDyvBRpt/e7x9SBoCzgLuBj9G7MOAN9LLdA7xrSBk+CHxnmmn/ekgZwG3Rz+fIhJHcFiOxm0iStHej8spAkrQXloEkyTKQJI14GSRZneQVo56h5Vib5LhRz9CVHF3I0HIs+L/PLmToSo75zDAqp5ZO5/3ALyf5v1X19hHOAHAc8EtJFlXVqSOcoSs5upABuvHvswsZupJj3jJ4NhGQ5KVV9d1RzyBNpwv/PruQoSs55iPDyOwmatcbeXuS/9C+3j5+nZFh/WK7kGFvkvzaqGVI8rIkPzfF+C+PUoa2vpe3yy6QZCzJbyR5NQz1ObLgGabJ9V/Hby9UjvnOMBKvDJKcBWwAvgDsaMNLgV8DPlZVl49Chn1J8o2qOmpUMiQ5g961iXYCLwbeXVW3t2lfqarprg+zX2Vo6/pt4Dx6lz/4JL13294DvB74b1V12ShkaDkunjxE702IlwNU1bn7Y4ZROWbwn4FjqurJ/sEkh9K7WuUw/iPuQgaSTPchKaH3rsuRyND8Lr3fyaNJjgWuSPKRqvrfTH1NmP01A8D7gFcDLwEeBl5ZVY+1f583A8P4j7gLGaD3zu9b6P3hNv47OJPeB90My9AzjEoZhKkvBPZDhveE60IG6H2E3juB/zdpPMCxI5QB4ICqehSgqr6c5I3A55IsY+rf1f6aAeAHVfU08HSSfxi/WmdVPZFkWDm6kAFgFXA+sAb4j1X1zSQbqmrT/pxhVMrgAuArSb7AxIWmjqK3i+b8EcoAcCvwdFXdMnlCkgdGKAPAd5P8XFX9A0D76/wNwF/Q+wt1VDIAVJIXV9UPgNPGB9P7LOBhHVvsQobx/fEfTHIMcGWS64a5/oXKMBLHDOC53TGn0PtcUejtt7+hqp4YpQyakORX6JXSg5PGXwycUVVXjkKGtr6jgG9W1Z5J40uAX6yqvxyFDFNkCvBe4HVV9c5hr3+YGUaiDJKk9vGDPp95XugZupKjCxm6kqMLGbqSowsZupJjITKMyqmlNyd5f/vL4zlJDkxyYpJN9K5hv79n6EqOLmToSo4uZOhKji5k6EqOoWcYlVcGBwH/FngHsAJ4kt4ZCy+id7T+M1U1rx9m0oUMe8lxEHDAsHJ0IUNXcnQhQ1dydCHDXnLs9/9fjEQZ9Gv7Yo8Avjf5NM9RytCVHF3I0JUcXcjQlRxdyNCVHMPKMHJlIEn6caNyzECStBeWgSTJMtBoSHJukvuTTHnefpJ3J/kf00yb/E5pab8zKu9Alt4LvKmqti90kLmS3mcd7Nn3nNK++cpA+70kfwD8LHB9kg8l+YskdyW5NVNcJjrJiiR/m+TuJJ/Yy3LPasv5+yRXtLFfT3JbkjuT/GWSxW38o0k2JvlikoeSnLuP5Ywl+bMkt7evE/qWc0WSvwGumNMNpZHmKwPt96rqd5KsAd5I7zLid1bV6UlOpHe12NdOesingUuq6vIk50y1zPSusf9fgH9eVY8nOaxN+hJwfFVVkt8CPgx8qE37hZbhpcADSS4Bfn6a5XwauKiqvtTeeHQD8Itt2irg9VX1vVlvFGkSy0Cj5vXAvwGoqpuSHJ7kZZPmOWF8Hnp/fX9yiuWcCPxpVT3elrW7jS8FPpvkSOBA4Gt9j7muqp4BnkmyE1i8l+W8CViVPHdB25cl+el2e7NFoLlmGUhTm+0bcP478Kmq2pze1Uc/2jftmb7bz7L359+L6L3C+H7/YCuHf5xlNmlaHjPQqPlrem/xp/1n/XhVfWfSPH9D74NEGJ93XJKvtps3AW9LcngbH9+9czATn2T3fK4dM91yvkDvw8/H1/va57EsadYsA42ajwLHJLkLuJCp/8P+AHBOkruZuNw4SY6gfRBRVd1L7zMqbkny98Cn+pb/p0nuAB7fV5i9LOdcYHU7sHwf8Dsz/DmlGfFyFNLzlOQtwM9W1eTPp5Ve8CwDSZK7iSRJloEkCctAkoRlIEnCMpAkYRlIkrAMJEnA/wcgKqYskpoZFwAAAABJRU5ErkJggg==\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":"# Dataset","metadata":{}},{"cell_type":"code","source":"def read_img_and_cvt_format(img_path, clr_format=cv2.COLOR_BGR2RGB):\n    return cv2.cvtColor(cv2.imread(img_path), clr_format)\n\nclass RSNADataset(Dataset):\n    \n    def __init__(self, df, is_test=False, transforms=None):\n        super(RSNADataset, self).__init__()\n        self.df = df\n        self.is_test = is_test\n        self.transforms = transforms\n            \n    def __len__(self):\n        return len(self.df)\n    \n    def __getitem__(self, idx):\n        row = self.df.iloc[idx]\n        img_path = BASE_IMG_DIR + f\"{row.patient_id}_{row.image_id}.png\"\n        img = read_img_and_cvt_format(img_path)\n        if self.transforms:\n            img = self.transforms(image=img)[\"image\"]\n            \n        label = -1\n        if not self.is_test:\n            label = torch.tensor(row.cancer, dtype=torch.float32).float()\n\n        return img, label\n            ","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:38.086045Z","iopub.execute_input":"2023-03-14T01:56:38.087093Z","iopub.status.idle":"2023-03-14T01:56:38.095795Z","shell.execute_reply.started":"2023-03-14T01:56:38.087054Z","shell.execute_reply":"2023-03-14T01:56:38.094673Z"},"trusted":true},"execution_count":14,"outputs":[]},{"cell_type":"markdown","source":"# Augmentations","metadata":{}},{"cell_type":"code","source":"def _get_train_transforms_without_aug():\n    return albumentations.Compose([\n        albumentations.Normalize(\n            mean=[0.485, 0.456, 0.406], \n            std=[0.229, 0.224, 0.225]\n        ),\n        ToTensorV2()\n    ])","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:38.09714Z","iopub.execute_input":"2023-03-14T01:56:38.097583Z","iopub.status.idle":"2023-03-14T01:56:38.107624Z","shell.execute_reply.started":"2023-03-14T01:56:38.097539Z","shell.execute_reply":"2023-03-14T01:56:38.106661Z"},"trusted":true},"execution_count":15,"outputs":[]},{"cell_type":"code","source":"train_transforms = _get_train_transforms_without_aug()\ndataset = RSNADataset(df_all, is_test=False, transforms=train_transforms)\ndata_loader = DataLoader(dataset, batch_size=2)\n# plt.imshow(dataset[0][0], cmap=\"bone\")","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:38.108957Z","iopub.execute_input":"2023-03-14T01:56:38.109406Z","iopub.status.idle":"2023-03-14T01:56:38.118123Z","shell.execute_reply.started":"2023-03-14T01:56:38.109371Z","shell.execute_reply":"2023-03-14T01:56:38.117226Z"},"trusted":true},"execution_count":16,"outputs":[]},{"cell_type":"markdown","source":"# Model","metadata":{}},{"cell_type":"code","source":"class RSNAModel(nn.Module):\n    \n    def __init__(self, model_name, pretrained=True):\n        super(RSNAModel, self).__init__()\n        self.model = timm.create_model(model_name, pretrained=pretrained)\n        if \"efficientnet\" in CFG.model_name:\n            in_features = self.model.classifier.in_features\n            self.model.classifier = nn.Linear(in_features, CFG.n_classes)\n        elif \"resnet\" in CFG.model_name:\n            in_features = self.model.fc.in_features\n            self.model.fc = nn.Linear(in_features, CFG.n_classes)\n\n    def forward(self, img):\n        return self.model(img)","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:38.119458Z","iopub.execute_input":"2023-03-14T01:56:38.119919Z","iopub.status.idle":"2023-03-14T01:56:38.129859Z","shell.execute_reply.started":"2023-03-14T01:56:38.119881Z","shell.execute_reply":"2023-03-14T01:56:38.128801Z"},"trusted":true},"execution_count":17,"outputs":[]},{"cell_type":"markdown","source":"# Utilities","metadata":{}},{"cell_type":"code","source":"LOGS_PATH = Path(\"logs\")\nLOGS_PATH.mkdir(exist_ok=True)\n\nclass AverageMeter:\n    \n    def __init__(self):\n        self.reset()\n    \n    def reset(self):\n        self.val = 0\n        self.avg = 0\n        self.sum = 0\n        self.count = 0\n    \n    def update(self, val, n=1):\n        self.val = val\n        self.sum += val * n\n        self.count += n\n        self.avg = self.sum / self.count\n        \n        \ndef as_minutes(s):\n    m = math.floor(s / 60)\n    s -= m * 60\n    return f\"{m}m {s}s\"\n\n\ndef time_since(since, percent):\n    now = time.time()\n    s = now - since\n    es = s / percent\n    rs = es - s\n    return f\"{as_minutes(s)} (remain {as_minutes(rs)})\"\n\n\ndef init_logger(log_file=LOGS_PATH / 'train.log'):\n    from logging import getLogger, INFO, FileHandler,  Formatter,  StreamHandler\n    logger = getLogger(__name__)\n    logger.setLevel(INFO)\n    handler1 = StreamHandler()\n    handler1.setFormatter(Formatter(\"%(message)s\"))\n    handler2 = FileHandler(filename=log_file)\n    handler2.setFormatter(Formatter(\"%(message)s\"))\n    logger.addHandler(handler1)\n    logger.addHandler(handler2)\n    return logger\n\n\nLOGGER = init_logger()","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:38.133664Z","iopub.execute_input":"2023-03-14T01:56:38.134013Z","iopub.status.idle":"2023-03-14T01:56:38.146982Z","shell.execute_reply.started":"2023-03-14T01:56:38.133977Z","shell.execute_reply":"2023-03-14T01:56:38.146058Z"},"trusted":true},"execution_count":18,"outputs":[]},{"cell_type":"markdown","source":"# Train and Evaluation Steps","metadata":{}},{"cell_type":"code","source":"def train_step(model, data_loader, criterion, optimizer, epoch, scheduler, device):\n    \"\"\"\n    There is no scheduler update currently.\n    \"\"\"\n    batch_time = AverageMeter()\n    data_time = AverageMeter()\n    losses = AverageMeter()\n    # scores = AverageMeter()\n    \n    model.train()\n    start = end = time.time()\n    # global_step = 0\n    total_len = len(data_loader)\n    \n    for step, (images, labels) in enumerate(data_loader):\n        \n        data_time.update(time.time() - end)\n        images = images.to(device)\n        labels = labels.to(device)\n        batch_size = labels.size(0)\n        preds = model(images).squeeze()\n        loss = criterion(preds, labels)\n        losses.update(loss.item(), batch_size)\n        \n        if CFG.gradient_accumulation_steps > 1:\n            loss = loss / CFG.gradient_accumulation_steps\n        \n        loss.backward()\n        grad_norm = torch.nn.utils.clip_grad_norm_(model.parameters(), \n                                                   CFG.max_grad_norm)\n        if (step + 1) % CFG.gradient_accumulation_steps == 0:\n            optimizer.step()\n            optimizer.zero_grad()\n            # global_step += 1\n        \n        batch_time.update(time.time() - end)\n        end = time.time()\n        if step % CFG.print_every == 0 or step == (total_len - 1):\n            print(f\"Epoch: [{epoch+1}][{step}/{total_len}] \"\n                  f\"Data: {data_time.val:.3f} ({data_time.avg:.3f}) \"\n                  f\"Batch: {batch_time.val:.3f} ({batch_time.avg:.3f}) \"\n                  f\"Elapsed: {time_since(start, float(step + 1) / (total_len))} \"\n                  f\"Loss: {losses.val:.5f}({losses.avg:.5f}) \"\n                  f\"Grad: {grad_norm:.4f}\" # LR: {lr:.6f}\n                 )\n    \n    return losses.avg\n            \n\ndef valid_step(model, data_loader, criterion, device):\n    \n    batch_time = AverageMeter()\n    data_time = AverageMeter()\n    losses = AverageMeter()\n    scores = AverageMeter()\n    \n    model.eval()\n    start = end = time.time()\n    total_len = len(data_loader)\n    predictions = []\n    \n    for step, (images, labels) in enumerate(data_loader):\n        data_time.update(time.time() - end)\n        images = images.to(device)\n        labels = labels.to(device)\n        batch_size = labels.size(0)\n        \n        with torch.no_grad():\n            preds = model(images).squeeze()\n        \n        loss = criterion(preds, labels)\n        losses.update(loss.item(), batch_size)\n        predictions.append(preds.sigmoid().cpu().numpy())\n        \n        if CFG.gradient_accumulation_steps > 1:\n            loss = loss / CFG.gradient_accumulation_steps\n            \n        batch_time.update(time.time() - end)\n        end = time.time()\n        \n        if step % CFG.print_every == 0 or step == (total_len - 1):\n            print(f\"Eval: [{step}/{total_len}] \"\n                  f\"Data: {data_time.val:.3f} ({data_time.avg:.3f}) \"\n                  f\"Batch: {batch_time.val:.3f} ({batch_time.avg:.3f}) \"\n                  f\"Elapsed: {time_since(start, float(step + 1) / total_len)} \"\n                  f\"Loss: {losses.val:.5f} ({losses.avg:.5f})\"\n                 )\n    \n    predictions = np.concatenate(predictions)\n    return losses.avg, predictions","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:38.148413Z","iopub.execute_input":"2023-03-14T01:56:38.148965Z","iopub.status.idle":"2023-03-14T01:56:38.165128Z","shell.execute_reply.started":"2023-03-14T01:56:38.148929Z","shell.execute_reply":"2023-03-14T01:56:38.164161Z"},"trusted":true},"execution_count":19,"outputs":[]},{"cell_type":"code","source":"def pfbeta(labels, predictions, beta):\n    \"\"\"\n    from here: https://www.kaggle.com/code/sohier/probabilistic-f-score\n    \"\"\"\n    y_true_count = 0\n    ctp = 0\n    cfp = 0\n\n    for idx in range(len(labels)):\n        prediction = min(max(predictions[idx], 0), 1)\n        if (labels[idx]):\n            y_true_count += 1\n            ctp += prediction\n            cfp += 1 - prediction\n        else:\n            cfp += prediction\n\n    beta_squared = beta * beta\n    c_precision = ctp / (ctp + cfp)\n    c_recall = ctp / y_true_count\n    if (c_precision > 0 and c_recall > 0):\n        result = (1 + beta_squared) * (c_precision * c_recall) / (beta_squared * c_precision + c_recall)\n        return result\n    else:\n        return 0\n","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:38.166602Z","iopub.execute_input":"2023-03-14T01:56:38.166982Z","iopub.status.idle":"2023-03-14T01:56:38.180777Z","shell.execute_reply.started":"2023-03-14T01:56:38.166908Z","shell.execute_reply":"2023-03-14T01:56:38.1799Z"},"trusted":true},"execution_count":20,"outputs":[]},{"cell_type":"code","source":"MODELS_DIR = Path(\"models\")\nMODELS_DIR.mkdir(exist_ok=False)","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:38.182354Z","iopub.execute_input":"2023-03-14T01:56:38.182703Z","iopub.status.idle":"2023-03-14T01:56:38.19141Z","shell.execute_reply.started":"2023-03-14T01:56:38.18267Z","shell.execute_reply":"2023-03-14T01:56:38.190417Z"},"trusted":true},"execution_count":21,"outputs":[]},{"cell_type":"markdown","source":"# Full Training","metadata":{}},{"cell_type":"code","source":"device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\noof_df = pd.DataFrame()\n\nfor fold in range(CFG.n_folds):\n    LOGGER.info(f\"===== STARTING FOLD {fold}: ======\")\n    df_train = df_all[df_all.fold != fold].reset_index(drop=True)\n    df_valid = df_all[df_all.fold == fold].reset_index(drop=True)\n    \n    train_dataset = RSNADataset(df_train, transforms=train_transforms)\n    valid_dataset = RSNADataset(df_valid, transforms=train_transforms)\n    train_data_loader = DataLoader(train_dataset, batch_size=CFG.train_batch_size, shuffle=True)\n    valid_data_loader = DataLoader(valid_dataset, batch_size=CFG.valid_batch_size, shuffle=False)\n    \n    model = RSNAModel(CFG.model_name)\n    model.to(device)\n    optimizer = optim.Adam(model.parameters(), \n                           lr=CFG.lr, \n                           weight_decay=CFG.wd)\n    criterion = nn.BCEWithLogitsLoss()\n    best_score = 0.0\n    best_loss = np.inf\n\n    for epoch in range(CFG.n_epochs):\n        \n        start_time = time.time()\n        avg_epoch_loss = train_step(model, \n                                    train_data_loader, \n                                    criterion, \n                                    optimizer, \n                                    epoch, \n                                    scheduler=None, \n                                    device=device)\n\n        avg_valid_loss, valid_preds = valid_step(model, \n                                                 valid_data_loader, \n                                                 criterion, \n                                                 device)\n        score = pfbeta(df_valid.cancer, valid_preds, beta=1)\n        elapsed = time.time() - start_time\n        LOGGER.info(f\"Epoch: {epoch+1} - avg_epoch_loss: {avg_epoch_loss:.5f} - avg_val_loss: {avg_valid_loss:.5f} - time: {elapsed:.0f}s\")\n        \n        if score > best_score:\n            best_score = score\n            LOGGER.info(f\"Epoch: {epoch+1} - Save best score: {best_score:.4f}\")\n            torch.save({\n                \"model\": model.state_dict(),\n                \"preds\": valid_preds\n            }, str(MODELS_DIR / f\"{CFG.model_name}_fold_{fold}_best.pth\"))\n\n    check_point = torch.load(str(MODELS_DIR / f\"{CFG.model_name}_fold_{fold}_best.pth\"))\n    df_tmp = pd.DataFrame()\n    df_tmp[\"labels\"] = df_valid.cancer\n    df_tmp[\"preds\"] = check_point[\"preds\"]\n    df_tmp[\"fold\"] = fold\n    oof_df = pd.concat([oof_df, df_tmp])","metadata":{"execution":{"iopub.status.busy":"2023-03-14T01:56:38.192931Z","iopub.execute_input":"2023-03-14T01:56:38.193287Z","iopub.status.idle":"2023-03-14T05:36:50.749955Z","shell.execute_reply.started":"2023-03-14T01:56:38.193253Z","shell.execute_reply":"2023-03-14T05:36:50.74886Z"},"trusted":true},"execution_count":22,"outputs":[{"name":"stderr","text":"===== STARTING FOLD 0: ======\nDownloading: \"https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50d_ra2-464e36ba.pth\" to /root/.cache/torch/hub/checkpoints/resnet50d_ra2-464e36ba.pth\n","output_type":"stream"},{"name":"stdout","text":"Epoch: [1][0/684] Data: 0.953 (0.953) Batch: 7.518 (7.518) Elapsed: 0m 7.517764568328857s (remain 85m 34.63320016860962s) Loss: 0.74088(0.74088) Grad: 3.7501\nEpoch: [1][100/684] Data: 0.817 (0.933) Batch: 1.286 (1.465) Elapsed: 2m 27.94307279586792s (remain 14m 13.968430098920749s) Loss: 0.09349(0.15540) Grad: 1.3885\nEpoch: [1][200/684] Data: 0.832 (0.887) Batch: 1.302 (1.388) Elapsed: 4m 38.89942526817322s (remain 11m 10.191156241431031s) Loss: 0.09298(0.13076) Grad: 1.5594\nEpoch: [1][300/684] Data: 0.774 (0.862) Batch: 1.243 (1.353) Elapsed: 6m 47.11553144454956s (remain 8m 38.024081539078s) Loss: 0.09036(0.12343) Grad: 0.9839\nEpoch: [1][400/684] Data: 0.778 (0.849) Batch: 1.247 (1.334) Elapsed: 8m 55.04655361175537s (remain 6m 17.60143309757302s) Loss: 0.08119(0.11818) Grad: 0.8656\nEpoch: [1][500/684] Data: 0.862 (0.842) Batch: 1.331 (1.324) Elapsed: 11m 3.4908621311187744s (remain 4m 2.352949640708175s) Loss: 0.01963(0.11615) Grad: 0.2484\nEpoch: [1][600/684] Data: 0.843 (0.841) Batch: 1.312 (1.321) Elapsed: 13m 14.19685173034668s (remain 1m 49.681095996037925s) Loss: 0.16012(0.11368) Grad: 0.8607\nEpoch: [1][683/684] Data: 0.785 (0.843) Batch: 1.232 (1.322) Elapsed: 15m 4.110965967178345s (remain 0m 0.0s) Loss: 0.07407(0.11049) Grad: 1.1927\nEval: [0/171] Data: 0.798 (0.798) Batch: 0.938 (0.938) Elapsed: 0m 0.9376852512359619s (remain 2m 39.406492710113525s) Loss: 0.22969 (0.22969)\nEval: [100/171] Data: 0.920 (0.833) Batch: 1.062 (0.973) Elapsed: 1m 38.26230072975159s (remain 1m 8.102584664184263s) Loss: 0.01774 (0.09827)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 1 - avg_epoch_loss: 0.11049 - avg_val_loss: 0.10127 - time: 1070s\nEpoch: 1 - Save best score: 0.0181\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.632 (0.828) Batch: 0.761 (0.968) Elapsed: 2m 45.48296332359314s (remain 0m 0.0s) Loss: 0.16620 (0.10127)\nEpoch: [2][0/684] Data: 0.181 (0.181) Batch: 0.650 (0.650) Elapsed: 0m 0.6503791809082031s (remain 7m 24.208980560302734s) Loss: 0.13013(0.13013) Grad: 1.0037\nEpoch: [2][100/684] Data: 0.187 (0.191) Batch: 0.657 (0.661) Elapsed: 1m 6.806338548660278s (remain 6m 25.624706671969705s) Loss: 0.01983(0.08823) Grad: 0.2362\nEpoch: [2][200/684] Data: 0.173 (0.188) Batch: 0.642 (0.658) Elapsed: 2m 12.347289800643921s (remain 5m 18.02856205826373s) Loss: 0.09106(0.08811) Grad: 0.8677\nEpoch: [2][300/684] Data: 0.175 (0.187) Batch: 0.644 (0.657) Elapsed: 3m 17.74503183364868s (remain 4m 11.615771402948326s) Loss: 0.02038(0.08571) Grad: 0.3374\nEpoch: [2][400/684] Data: 0.203 (0.186) Batch: 0.673 (0.656) Elapsed: 4m 23.061318159103394s (remain 3m 5.651753214529322s) Loss: 0.10236(0.08522) Grad: 1.0811\nEpoch: [2][500/684] Data: 0.173 (0.186) Batch: 0.642 (0.656) Elapsed: 5m 28.70646381378174s (remain 2m 0.06643289006404984s) Loss: 0.25674(0.08697) Grad: 2.4299\nEpoch: [2][600/684] Data: 0.180 (0.186) Batch: 0.649 (0.656) Elapsed: 6m 34.45998740196228s (remain 0m 54.47617130509627s) Loss: 0.14821(0.08797) Grad: 1.4037\nEpoch: [2][683/684] Data: 0.176 (0.187) Batch: 0.623 (0.657) Elapsed: 7m 29.275219440460205s (remain 0m 0.0s) Loss: 0.05861(0.08857) Grad: 0.6974\nEval: [0/171] Data: 0.180 (0.180) Batch: 0.320 (0.320) Elapsed: 0m 0.3199799060821533s (remain 0m 54.396584033966064s) Loss: 0.16433 (0.16433)\nEval: [100/171] Data: 0.188 (0.185) Batch: 0.327 (0.325) Elapsed: 0m 32.82875180244446s (remain 0m 22.752600259119916s) Loss: 0.01792 (0.10110)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 2 - avg_epoch_loss: 0.08857 - avg_val_loss: 0.10302 - time: 505s\nEpoch: 2 - Save best score: 0.0230\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.148 (0.185) Batch: 0.277 (0.325) Elapsed: 0m 55.541526079177856s (remain 0m 0.0s) Loss: 0.12959 (0.10302)\nEpoch: [3][0/684] Data: 0.206 (0.206) Batch: 0.676 (0.676) Elapsed: 0m 0.6758205890655518s (remain 7m 41.58546233177185s) Loss: 0.07439(0.07439) Grad: 0.6447\nEpoch: [3][100/684] Data: 0.204 (0.187) Batch: 0.673 (0.657) Elapsed: 1m 6.361910343170166s (remain 6m 23.05934386206144s) Loss: 0.06532(0.05969) Grad: 1.4477\nEpoch: [3][200/684] Data: 0.173 (0.186) Batch: 0.642 (0.656) Elapsed: 2m 11.829660415649414s (remain 5m 16.784706371933623s) Loss: 0.13407(0.05774) Grad: 3.9772\nEpoch: [3][300/684] Data: 0.194 (0.185) Batch: 0.664 (0.655) Elapsed: 3m 17.26664686203003s (remain 4m 11.007062286237556s) Loss: 0.01665(0.05679) Grad: 0.4132\nEpoch: [3][400/684] Data: 0.178 (0.185) Batch: 0.647 (0.655) Elapsed: 4m 22.65165686607361s (remain 3m 5.362640631169143s) Loss: 0.01692(0.05701) Grad: 0.5637\nEpoch: [3][500/684] Data: 0.177 (0.185) Batch: 0.646 (0.655) Elapsed: 5m 28.18060541152954s (remain 1m 59.87435287487011s) Loss: 0.04978(0.05774) Grad: 1.7664\nEpoch: [3][600/684] Data: 0.172 (0.185) Batch: 0.641 (0.655) Elapsed: 6m 33.63363695144653s (remain 0m 54.3620496954577s) Loss: 0.05958(0.05864) Grad: 1.6380\nEpoch: [3][683/684] Data: 0.189 (0.185) Batch: 0.636 (0.655) Elapsed: 7m 27.744470834732056s (remain 0m 0.0s) Loss: 0.01739(0.06098) Grad: 0.5888\nEval: [0/171] Data: 0.173 (0.173) Batch: 0.312 (0.312) Elapsed: 0m 0.31216859817504883s (remain 0m 53.0686616897583s) Loss: 0.14052 (0.14052)\nEval: [100/171] Data: 0.183 (0.182) Batch: 0.322 (0.322) Elapsed: 0m 32.495891094207764s (remain 0m 22.521904718757852s) Loss: 0.01880 (0.11848)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 3 - avg_epoch_loss: 0.06098 - avg_val_loss: 0.12193 - time: 503s\nEpoch: 3 - Save best score: 0.0417\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.142 (0.183) Batch: 0.271 (0.323) Elapsed: 0m 55.16566348075867s (remain 0m 0.0s) Loss: 0.16626 (0.12193)\nEpoch: [4][0/684] Data: 0.178 (0.178) Batch: 0.648 (0.648) Elapsed: 0m 0.6477947235107422s (remain 7m 22.443796157836914s) Loss: 0.01457(0.01457) Grad: 0.3191\nEpoch: [4][100/684] Data: 0.171 (0.186) Batch: 0.640 (0.656) Elapsed: 1m 6.280354022979736s (remain 6m 22.588578172249356s) Loss: 0.00591(0.02091) Grad: 0.2417\nEpoch: [4][200/684] Data: 0.172 (0.183) Batch: 0.641 (0.653) Elapsed: 2m 11.247389316558838s (remain 5m 15.38551761143242s) Loss: 0.01150(0.01980) Grad: 1.4842\nEpoch: [4][300/684] Data: 0.173 (0.183) Batch: 0.642 (0.653) Elapsed: 3m 16.59723424911499s (remain 4m 10.155284775451946s) Loss: 0.03189(0.02059) Grad: 2.2014\nEpoch: [4][400/684] Data: 0.187 (0.185) Batch: 0.658 (0.655) Elapsed: 4m 22.660145044326782s (remain 3m 5.368631041258027s) Loss: 0.00614(0.01939) Grad: 0.5883\nEpoch: [4][500/684] Data: 0.182 (0.186) Batch: 0.651 (0.656) Elapsed: 5m 28.83770251274109s (remain 2m 0.11437037890544843s) Loss: 0.05200(0.02012) Grad: 3.1516\nEpoch: [4][600/684] Data: 0.207 (0.186) Batch: 0.677 (0.656) Elapsed: 6m 34.33467984199524s (remain 0m 54.45886593491781s) Loss: 0.00723(0.02043) Grad: 1.0356\nEpoch: [4][683/684] Data: 0.224 (0.186) Batch: 0.673 (0.656) Elapsed: 7m 28.72268271446228s (remain 0m 0.0s) Loss: 0.01086(0.02099) Grad: 0.7452\nEval: [0/171] Data: 0.173 (0.173) Batch: 0.312 (0.312) Elapsed: 0m 0.3121926784515381s (remain 0m 53.072755336761475s) Loss: 0.64266 (0.64266)\nEval: [100/171] Data: 0.190 (0.205) Batch: 0.330 (0.344) Elapsed: 0m 34.775309801101685s (remain 0m 24.10169986214968s) Loss: 0.00145 (0.17235)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 4 - avg_epoch_loss: 0.02099 - avg_val_loss: 0.18098 - time: 506s\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.144 (0.196) Batch: 0.273 (0.336) Elapsed: 0m 57.43510460853577s (remain 0m 0.0s) Loss: 0.27634 (0.18098)\nEpoch: [5][0/684] Data: 0.169 (0.169) Batch: 0.639 (0.639) Elapsed: 0m 0.6386699676513672s (remain 7m 16.21158790588379s) Loss: 0.00183(0.00183) Grad: 0.1029\nEpoch: [5][100/684] Data: 0.171 (0.184) Batch: 0.641 (0.654) Elapsed: 1m 6.053454637527466s (remain 6m 21.27885201661894s) Loss: 0.00107(0.01086) Grad: 0.0794\nEpoch: [5][200/684] Data: 0.175 (0.186) Batch: 0.645 (0.655) Elapsed: 2m 11.729381084442139s (remain 5m 16.543736635748985s) Loss: 0.00801(0.01025) Grad: 1.4325\nEpoch: [5][300/684] Data: 0.173 (0.186) Batch: 0.642 (0.656) Elapsed: 3m 17.456830501556396s (remain 4m 11.249056751149794s) Loss: 0.00590(0.00940) Grad: 0.6856\nEpoch: [5][400/684] Data: 0.202 (0.186) Batch: 0.671 (0.655) Elapsed: 4m 22.842355966567993s (remain 3m 5.497223786879658s) Loss: 0.09626(0.00938) Grad: 5.1202\nEpoch: [5][500/684] Data: 0.202 (0.185) Batch: 0.671 (0.655) Elapsed: 5m 28.313739776611328s (remain 1m 59.922982792654466s) Loss: 0.00180(0.01027) Grad: 0.1938\nEpoch: [5][600/684] Data: 0.210 (0.185) Batch: 0.678 (0.655) Elapsed: 6m 33.76538705825806s (remain 0m 54.380244801722824s) Loss: 0.00076(0.01038) Grad: 0.0726\nEpoch: [5][683/684] Data: 0.228 (0.186) Batch: 0.678 (0.656) Elapsed: 7m 28.494950771331787s (remain 0m 0.0s) Loss: 0.00139(0.01095) Grad: 0.1190\nEval: [0/171] Data: 0.186 (0.186) Batch: 0.326 (0.326) Elapsed: 0m 0.3260054588317871s (remain 0m 55.42092800140381s) Loss: 0.69810 (0.69810)\nEval: [100/171] Data: 0.192 (0.180) Batch: 0.335 (0.320) Elapsed: 0m 32.2760055065155s (remain 0m 22.369508766891933s) Loss: 0.00049 (0.20302)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 5 - avg_epoch_loss: 0.01095 - avg_val_loss: 0.21172 - time: 504s\n===== STARTING FOLD 1: ======\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.161 (0.182) Batch: 0.290 (0.321) Elapsed: 0m 54.925025939941406s (remain 0m 0.0s) Loss: 0.32665 (0.21172)\nEpoch: [1][0/684] Data: 0.173 (0.173) Batch: 0.643 (0.643) Elapsed: 0m 0.6432695388793945s (remain 7m 19.353095054626465s) Loss: 0.76014(0.76014) Grad: 4.1440\nEpoch: [1][100/684] Data: 0.175 (0.184) Batch: 0.644 (0.654) Elapsed: 1m 6.014967918395996s (remain 6m 21.05669600420657s) Loss: 0.16207(0.13983) Grad: 1.7960\nEpoch: [1][200/684] Data: 0.233 (0.184) Batch: 0.704 (0.654) Elapsed: 2m 11.36206340789795s (remain 5m 15.661077741366682s) Loss: 0.08625(0.12251) Grad: 1.0446\nEpoch: [1][300/684] Data: 0.180 (0.184) Batch: 0.652 (0.654) Elapsed: 3m 16.880452394485474s (remain 4m 10.515658694644287s) Loss: 0.19679(0.11685) Grad: 1.3159\nEpoch: [1][400/684] Data: 0.261 (0.187) Batch: 0.731 (0.656) Elapsed: 4m 23.22107243537903s (remain 3m 5.764497504269968s) Loss: 0.16760(0.10976) Grad: 1.5147\nEpoch: [1][500/684] Data: 0.270 (0.187) Batch: 0.740 (0.657) Elapsed: 5m 29.169084072113037s (remain 2m 0.2354139425083872s) Loss: 0.20684(0.10928) Grad: 0.8906\nEpoch: [1][600/684] Data: 0.213 (0.187) Batch: 0.687 (0.657) Elapsed: 6m 35.004631996154785s (remain 0m 54.55138844539243s) Loss: 0.19282(0.11096) Grad: 0.9022\nEpoch: [1][683/684] Data: 0.120 (0.188) Batch: 0.475 (0.657) Elapsed: 7m 29.628005027770996s (remain 0m 0.0s) Loss: 0.09736(0.11073) Grad: 0.6031\nEval: [0/172] Data: 0.229 (0.229) Batch: 0.369 (0.369) Elapsed: 0m 0.36904168128967285s (remain 1m 3.1061275005340576s) Loss: 0.13556 (0.13556)\nEval: [100/172] Data: 0.199 (0.187) Batch: 0.342 (0.327) Elapsed: 0m 32.986058712005615s (remain 0m 23.188219490617804s) Loss: 0.03329 (0.10037)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 1 - avg_epoch_loss: 0.11073 - avg_val_loss: 0.10299 - time: 506s\nEpoch: 1 - Save best score: 0.0184\n","output_type":"stream"},{"name":"stdout","text":"Eval: [171/172] Data: 0.018 (0.187) Batch: 0.039 (0.325) Elapsed: 0m 55.98339104652405s (remain 0m 0.0s) Loss: 0.02097 (0.10299)\nEpoch: [2][0/684] Data: 0.205 (0.205) Batch: 0.674 (0.674) Elapsed: 0m 0.6745069026947021s (remain 7m 40.68821454048157s) Loss: 0.02387(0.02387) Grad: 0.2879\nEpoch: [2][100/684] Data: 0.177 (0.187) Batch: 0.646 (0.656) Elapsed: 1m 6.286669731140137s (remain 6m 22.625034190640577s) Loss: 0.07030(0.08620) Grad: 1.3695\nEpoch: [2][200/684] Data: 0.219 (0.185) Batch: 0.689 (0.655) Elapsed: 2m 11.661370992660522s (remain 5m 16.380309400273745s) Loss: 0.01607(0.08857) Grad: 0.2008\nEpoch: [2][300/684] Data: 0.241 (0.187) Batch: 0.718 (0.656) Elapsed: 3m 17.5684335231781s (remain 4m 11.391063253744846s) Loss: 0.02088(0.08876) Grad: 0.2795\nEpoch: [2][400/684] Data: 0.245 (0.187) Batch: 0.716 (0.657) Elapsed: 4m 23.514518976211548s (remain 3m 5.971593192687919s) Loss: 0.07889(0.08879) Grad: 0.7515\nEpoch: [2][500/684] Data: 0.248 (0.188) Batch: 0.719 (0.657) Elapsed: 5m 29.360265493392944s (remain 2m 0.3052466772273874s) Loss: 0.11010(0.08945) Grad: 1.3526\nEpoch: [2][600/684] Data: 0.229 (0.188) Batch: 0.698 (0.658) Elapsed: 6m 35.63622975349426s (remain 0m 54.638614092412695s) Loss: 0.16854(0.08894) Grad: 2.0136\nEpoch: [2][683/684] Data: 0.124 (0.190) Batch: 0.468 (0.659) Elapsed: 7m 30.855669736862183s (remain 0m 0.0s) Loss: 0.05677(0.08920) Grad: 0.7526\nEval: [0/172] Data: 0.172 (0.172) Batch: 0.312 (0.312) Elapsed: 0m 0.3115267753601074s (remain 0m 53.27107858657837s) Loss: 0.08867 (0.08867)\nEval: [100/172] Data: 0.181 (0.196) Batch: 0.320 (0.336) Elapsed: 0m 33.90413737297058s (remain 0m 23.833601519612976s) Loss: 0.00999 (0.09897)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 2 - avg_epoch_loss: 0.08920 - avg_val_loss: 0.10374 - time: 508s\nEpoch: 2 - Save best score: 0.0241\n","output_type":"stream"},{"name":"stdout","text":"Eval: [171/172] Data: 0.017 (0.195) Batch: 0.038 (0.334) Elapsed: 0m 57.400113344192505s (remain 0m 0.0s) Loss: 0.00319 (0.10374)\nEpoch: [3][0/684] Data: 0.187 (0.187) Batch: 0.656 (0.656) Elapsed: 0m 0.6556715965270996s (remain 7m 27.823700428009033s) Loss: 0.05026(0.05026) Grad: 0.6373\nEpoch: [3][100/684] Data: 0.181 (0.198) Batch: 0.650 (0.667) Elapsed: 1m 7.38709831237793s (remain 6m 28.977013030854778s) Loss: 0.01972(0.05959) Grad: 0.4423\nEpoch: [3][200/684] Data: 0.171 (0.194) Batch: 0.640 (0.664) Elapsed: 2m 13.455559968948364s (remain 5m 20.69171873135349s) Loss: 0.04895(0.05540) Grad: 0.9148\nEpoch: [3][300/684] Data: 0.183 (0.191) Batch: 0.651 (0.661) Elapsed: 3m 18.95904517173767s (remain 4m 13.16051262716121s) Loss: 0.04427(0.05739) Grad: 1.5644\nEpoch: [3][400/684] Data: 0.179 (0.190) Batch: 0.648 (0.660) Elapsed: 4m 24.46526026725769s (remain 3m 6.642565226019769s) Loss: 0.08605(0.06004) Grad: 2.1815\nEpoch: [3][500/684] Data: 0.188 (0.188) Batch: 0.657 (0.658) Elapsed: 5m 29.63727068901062s (remain 2m 0.4064282157464163s) Loss: 0.03414(0.05997) Grad: 0.9290\nEpoch: [3][600/684] Data: 0.180 (0.188) Batch: 0.649 (0.658) Elapsed: 6m 35.33402991294861s (remain 0m 54.596879339059456s) Loss: 0.05200(0.06180) Grad: 1.1725\nEpoch: [3][683/684] Data: 0.124 (0.188) Batch: 0.469 (0.658) Elapsed: 7m 30.004047632217407s (remain 0m 0.0s) Loss: 0.12906(0.06197) Grad: 4.4995\nEval: [0/172] Data: 0.183 (0.183) Batch: 0.323 (0.323) Elapsed: 0m 0.3228604793548584s (remain 0m 55.209141969680786s) Loss: 0.09972 (0.09972)\nEval: [100/172] Data: 0.183 (0.184) Batch: 0.322 (0.324) Elapsed: 0m 32.730841875076294s (remain 0m 23.008809634954616s) Loss: 0.02959 (0.12094)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 3 - avg_epoch_loss: 0.06197 - avg_val_loss: 0.12992 - time: 505s\nEpoch: 3 - Save best score: 0.0423\n","output_type":"stream"},{"name":"stdout","text":"Eval: [171/172] Data: 0.019 (0.183) Batch: 0.039 (0.322) Elapsed: 0m 55.33070755004883s (remain 0m 0.0s) Loss: 0.02470 (0.12992)\nEpoch: [4][0/684] Data: 0.170 (0.170) Batch: 0.638 (0.638) Elapsed: 0m 0.6384832859039307s (remain 7m 16.084084272384644s) Loss: 0.01648(0.01648) Grad: 0.6306\nEpoch: [4][100/684] Data: 0.179 (0.184) Batch: 0.647 (0.654) Elapsed: 1m 6.016843795776367s (remain 6m 21.06752408849127s) Loss: 0.00946(0.02139) Grad: 0.5664\nEpoch: [4][200/684] Data: 0.171 (0.184) Batch: 0.640 (0.653) Elapsed: 2m 11.32169795036316s (remain 5m 15.564080149380118s) Loss: 0.00384(0.01804) Grad: 0.1996\nEpoch: [4][300/684] Data: 0.170 (0.184) Batch: 0.638 (0.653) Elapsed: 3m 16.686404943466187s (remain 4m 10.26874781843037s) Loss: 0.00616(0.01905) Grad: 0.5903\nEpoch: [4][400/684] Data: 0.178 (0.184) Batch: 0.646 (0.654) Elapsed: 4m 22.055460214614868s (remain 3m 4.9418833933566475s) Loss: 0.00373(0.02137) Grad: 0.1979\nEpoch: [4][500/684] Data: 0.172 (0.184) Batch: 0.642 (0.654) Elapsed: 5m 27.57023525238037s (remain 1m 59.65140329577969s) Loss: 0.00147(0.02167) Grad: 0.0874\nEpoch: [4][600/684] Data: 0.222 (0.184) Batch: 0.697 (0.654) Elapsed: 6m 32.88950824737549s (remain 0m 54.25928316893874s) Loss: 0.01050(0.02267) Grad: 1.3311\nEpoch: [4][683/684] Data: 0.130 (0.184) Batch: 0.474 (0.654) Elapsed: 7m 27.099188566207886s (remain 0m 0.0s) Loss: 0.00838(0.02321) Grad: 0.5649\nEval: [0/172] Data: 0.177 (0.177) Batch: 0.319 (0.319) Elapsed: 0m 0.31870293617248535s (remain 0m 54.498202085494995s) Loss: 0.12545 (0.12545)\nEval: [100/172] Data: 0.175 (0.185) Batch: 0.314 (0.324) Elapsed: 0m 32.7586886882782s (remain 0m 23.028385117502495s) Loss: 0.00621 (0.15833)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 4 - avg_epoch_loss: 0.02321 - avg_val_loss: 0.17161 - time: 503s\n","output_type":"stream"},{"name":"stdout","text":"Eval: [171/172] Data: 0.027 (0.183) Batch: 0.048 (0.322) Elapsed: 0m 55.381959199905396s (remain 0m 0.0s) Loss: 0.00004 (0.17161)\nEpoch: [5][0/684] Data: 0.248 (0.248) Batch: 0.718 (0.718) Elapsed: 0m 0.7177798748016357s (remain 8m 10.243654489517212s) Loss: 0.00498(0.00498) Grad: 0.4926\nEpoch: [5][100/684] Data: 0.234 (0.187) Batch: 0.703 (0.657) Elapsed: 1m 6.311976909637451s (remain 6m 22.77111424077856s) Loss: 0.00151(0.01077) Grad: 0.0975\nEpoch: [5][200/684] Data: 0.171 (0.185) Batch: 0.640 (0.655) Elapsed: 2m 11.617772817611694s (remain 5m 16.27554363635045s) Loss: 0.01921(0.00947) Grad: 4.7931\nEpoch: [5][300/684] Data: 0.183 (0.185) Batch: 0.652 (0.655) Elapsed: 3m 17.20073127746582s (remain 4m 10.923189632124263s) Loss: 0.00595(0.00964) Grad: 0.6394\nEpoch: [5][400/684] Data: 0.180 (0.186) Batch: 0.649 (0.656) Elapsed: 4m 22.988753080368042s (remain 3m 5.6005414507335445s) Loss: 0.02080(0.01208) Grad: 1.4316\nEpoch: [5][500/684] Data: 0.173 (0.187) Batch: 0.642 (0.657) Elapsed: 5m 29.009299516677856s (remain 2m 0.1770495240560308s) Loss: 0.01240(0.01247) Grad: 2.2529\nEpoch: [5][600/684] Data: 0.178 (0.187) Batch: 0.649 (0.657) Elapsed: 6m 34.66437911987305s (remain 0m 54.50439844750326s) Loss: 0.00062(0.01254) Grad: 0.0364\nEpoch: [5][683/684] Data: 0.119 (0.186) Batch: 0.463 (0.656) Elapsed: 7m 28.699369192123413s (remain 0m 0.0s) Loss: 0.00183(0.01282) Grad: 0.2617\nEval: [0/172] Data: 0.173 (0.173) Batch: 0.312 (0.312) Elapsed: 0m 0.31212401390075684s (remain 0m 53.37320637702942s) Loss: 0.04748 (0.04748)\nEval: [100/172] Data: 0.187 (0.182) Batch: 0.326 (0.321) Elapsed: 0m 32.447431325912476s (remain 0m 22.809580437027577s) Loss: 0.00474 (0.18595)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 5 - avg_epoch_loss: 0.01282 - avg_val_loss: 0.19590 - time: 504s\n===== STARTING FOLD 2: ======\n","output_type":"stream"},{"name":"stdout","text":"Eval: [171/172] Data: 0.019 (0.182) Batch: 0.039 (0.320) Elapsed: 0m 55.10779404640198s (remain 0m 0.0s) Loss: 0.00003 (0.19590)\nEpoch: [1][0/684] Data: 0.182 (0.182) Batch: 0.653 (0.653) Elapsed: 0m 0.6525628566741943s (remain 7m 25.70043110847473s) Loss: 0.68624(0.68624) Grad: 3.7897\nEpoch: [1][100/684] Data: 0.207 (0.184) Batch: 0.677 (0.654) Elapsed: 1m 6.0448102951049805s (remain 6m 21.228954475704995s) Loss: 0.19794(0.14281) Grad: 2.6368\nEpoch: [1][200/684] Data: 0.178 (0.185) Batch: 0.647 (0.655) Elapsed: 2m 11.587252616882324s (remain 5m 16.202204049523175s) Loss: 0.01715(0.12008) Grad: 0.2478\nEpoch: [1][300/684] Data: 0.199 (0.185) Batch: 0.668 (0.655) Elapsed: 3m 17.210137605667114s (remain 4m 10.935158481629571s) Loss: 0.02177(0.11588) Grad: 0.2673\nEpoch: [1][400/684] Data: 0.182 (0.186) Batch: 0.652 (0.656) Elapsed: 4m 22.899306774139404s (remain 3m 5.537416002696887s) Loss: 0.01755(0.11294) Grad: 0.2282\nEpoch: [1][500/684] Data: 0.175 (0.186) Batch: 0.645 (0.656) Elapsed: 5m 28.672653198242188s (remain 2m 0.05408290474719024s) Loss: 0.02729(0.11028) Grad: 0.3156\nEpoch: [1][600/684] Data: 0.185 (0.187) Batch: 0.654 (0.657) Elapsed: 6m 34.82299995422363s (remain 0m 54.52630448619061s) Loss: 0.07090(0.10905) Grad: 0.4480\nEpoch: [1][683/684] Data: 0.149 (0.187) Batch: 0.553 (0.657) Elapsed: 7m 29.371535301208496s (remain 0m 0.0s) Loss: 0.02812(0.10903) Grad: 0.3166\nEval: [0/171] Data: 0.173 (0.173) Batch: 0.312 (0.312) Elapsed: 0m 0.31217122077941895s (remain 0m 53.06910753250122s) Loss: 0.02494 (0.02494)\nEval: [100/171] Data: 0.179 (0.191) Batch: 0.319 (0.331) Elapsed: 0m 33.432656049728394s (remain 0m 23.1711477572375s) Loss: 0.02423 (0.09884)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 1 - avg_epoch_loss: 0.10903 - avg_val_loss: 0.10107 - time: 507s\nEpoch: 1 - Save best score: 0.0213\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.171 (0.194) Batch: 0.305 (0.334) Elapsed: 0m 57.091606855392456s (remain 0m 0.0s) Loss: 0.02392 (0.10107)\nEpoch: [2][0/684] Data: 0.178 (0.178) Batch: 0.652 (0.652) Elapsed: 0m 0.6517074108123779s (remain 7m 25.116161584854126s) Loss: 0.11379(0.11379) Grad: 0.5204\nEpoch: [2][100/684] Data: 0.246 (0.196) Batch: 0.716 (0.666) Elapsed: 1m 7.249888181686401s (remain 6m 28.18499811805117s) Loss: 0.09239(0.09016) Grad: 1.2035\nEpoch: [2][200/684] Data: 0.203 (0.192) Batch: 0.681 (0.662) Elapsed: 2m 12.969119548797607s (remain 5m 19.522809662036025s) Loss: 0.21641(0.08669) Grad: 2.2846\nEpoch: [2][300/684] Data: 0.172 (0.189) Batch: 0.641 (0.659) Elapsed: 3m 18.419040203094482s (remain 4m 12.47339667038267s) Loss: 0.27802(0.09067) Grad: 1.7385\nEpoch: [2][400/684] Data: 0.175 (0.188) Batch: 0.644 (0.658) Elapsed: 4m 24.043079137802124s (remain 3m 6.344616947625923s) Loss: 0.07026(0.08933) Grad: 0.5983\nEpoch: [2][500/684] Data: 0.182 (0.188) Batch: 0.651 (0.657) Elapsed: 5m 29.39344048500061s (remain 2m 0.317364488533201s) Loss: 0.02401(0.08994) Grad: 0.2671\nEpoch: [2][600/684] Data: 0.174 (0.187) Batch: 0.642 (0.657) Elapsed: 6m 34.93155264854431s (remain 0m 54.54129595645452s) Loss: 0.01697(0.08990) Grad: 0.2406\nEpoch: [2][683/684] Data: 0.163 (0.187) Batch: 0.565 (0.657) Elapsed: 7m 29.26575469970703s (remain 0m 0.0s) Loss: 0.03438(0.09152) Grad: 0.4638\nEval: [0/171] Data: 0.177 (0.177) Batch: 0.317 (0.317) Elapsed: 0m 0.3166825771331787s (remain 0m 53.83603811264038s) Loss: 0.03215 (0.03215)\nEval: [100/171] Data: 0.178 (0.183) Batch: 0.317 (0.323) Elapsed: 0m 32.60666465759277s (remain 0m 22.598678475559346s) Loss: 0.02392 (0.10327)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 2 - avg_epoch_loss: 0.09152 - avg_val_loss: 0.10447 - time: 505s\nEpoch: 2 - Save best score: 0.0293\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.167 (0.184) Batch: 0.301 (0.324) Elapsed: 0m 55.354512453079224s (remain 0m 0.0s) Loss: 0.04014 (0.10447)\nEpoch: [3][0/684] Data: 0.175 (0.175) Batch: 0.644 (0.644) Elapsed: 0m 0.644221305847168s (remain 7m 20.003151893615723s) Loss: 0.02838(0.02838) Grad: 0.3380\nEpoch: [3][100/684] Data: 0.178 (0.183) Batch: 0.647 (0.653) Elapsed: 1m 5.937964200973511s (remain 6m 20.612209199678773s) Loss: 0.06833(0.06055) Grad: 0.8524\nEpoch: [3][200/684] Data: 0.193 (0.184) Batch: 0.661 (0.654) Elapsed: 2m 11.402130603790283s (remain 5m 15.75735861507809s) Loss: 0.06103(0.06146) Grad: 1.4773\nEpoch: [3][300/684] Data: 0.174 (0.184) Batch: 0.643 (0.654) Elapsed: 3m 16.939332246780396s (remain 4m 10.590578905371729s) Loss: 0.01173(0.06421) Grad: 0.4557\nEpoch: [3][400/684] Data: 0.204 (0.186) Batch: 0.674 (0.656) Elapsed: 4m 23.05318832397461s (remain 3m 5.646015699962106s) Loss: 0.05651(0.06350) Grad: 1.1974\nEpoch: [3][500/684] Data: 0.171 (0.187) Batch: 0.641 (0.657) Elapsed: 5m 28.983309745788574s (remain 2m 0.16755625444977795s) Loss: 0.02378(0.06414) Grad: 0.5789\nEpoch: [3][600/684] Data: 0.180 (0.187) Batch: 0.650 (0.657) Elapsed: 6m 34.88704872131348s (remain 0m 54.53514982340937s) Loss: 0.09803(0.06558) Grad: 2.0024\nEpoch: [3][683/684] Data: 0.152 (0.187) Batch: 0.555 (0.657) Elapsed: 7m 29.47542929649353s (remain 0m 0.0s) Loss: 0.21767(0.06507) Grad: 3.7773\nEval: [0/171] Data: 0.178 (0.178) Batch: 0.317 (0.317) Elapsed: 0m 0.31731343269348145s (remain 0m 53.943283557891846s) Loss: 0.03536 (0.03536)\nEval: [100/171] Data: 0.170 (0.183) Batch: 0.309 (0.322) Elapsed: 0m 32.542734146118164s (remain 0m 22.554370200279912s) Loss: 0.01569 (0.13652)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 3 - avg_epoch_loss: 0.06507 - avg_val_loss: 0.13661 - time: 505s\nEpoch: 3 - Save best score: 0.0467\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.218 (0.184) Batch: 0.352 (0.323) Elapsed: 0m 55.26705312728882s (remain 0m 0.0s) Loss: 0.09308 (0.13661)\nEpoch: [4][0/684] Data: 0.177 (0.177) Batch: 0.646 (0.646) Elapsed: 0m 0.6460466384887695s (remain 7m 21.24985408782959s) Loss: 0.03115(0.03115) Grad: 1.2560\nEpoch: [4][100/684] Data: 0.258 (0.214) Batch: 0.728 (0.684) Elapsed: 1m 9.097107172012329s (remain 6m 38.84765823052658s) Loss: 0.00584(0.02272) Grad: 0.3586\nEpoch: [4][200/684] Data: 0.259 (0.202) Batch: 0.728 (0.672) Elapsed: 2m 15.050548315048218s (remain 5m 24.524451921235254s) Loss: 0.02407(0.02239) Grad: 1.9064\nEpoch: [4][300/684] Data: 0.256 (0.199) Batch: 0.726 (0.669) Elapsed: 3m 21.368629693984985s (remain 4m 16.22652881327656s) Loss: 0.05805(0.02121) Grad: 2.7514\nEpoch: [4][400/684] Data: 0.231 (0.197) Batch: 0.705 (0.667) Elapsed: 4m 27.45341205596924s (remain 3m 8.751410503339855s) Loss: 0.01631(0.02039) Grad: 1.0957\nEpoch: [4][500/684] Data: 0.229 (0.195) Batch: 0.699 (0.665) Elapsed: 5m 33.26528024673462s (remain 2m 1.7316293116815586s) Loss: 0.00559(0.02134) Grad: 0.4355\nEpoch: [4][600/684] Data: 0.401 (0.195) Batch: 0.893 (0.664) Elapsed: 6m 39.3643524646759s (remain 0m 55.153479624905344s) Loss: 0.01930(0.02304) Grad: 1.6986\nEpoch: [4][683/684] Data: 0.226 (0.195) Batch: 0.629 (0.664) Elapsed: 7m 34.39921498298645s (remain 0m 0.0s) Loss: 0.04900(0.02333) Grad: 4.4992\nEval: [0/171] Data: 0.194 (0.194) Batch: 0.334 (0.334) Elapsed: 0m 0.3341200351715088s (remain 0m 56.800405979156494s) Loss: 0.00605 (0.00605)\nEval: [100/171] Data: 0.179 (0.188) Batch: 0.318 (0.328) Elapsed: 0m 33.11618709564209s (remain 0m 22.951812838563825s) Loss: 0.00055 (0.17509)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 4 - avg_epoch_loss: 0.02333 - avg_val_loss: 0.16692 - time: 511s\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.175 (0.190) Batch: 0.309 (0.330) Elapsed: 0m 56.350708961486816s (remain 0m 0.0s) Loss: 0.02848 (0.16692)\nEpoch: [5][0/684] Data: 0.177 (0.177) Batch: 0.645 (0.645) Elapsed: 0m 0.6453127861022949s (remain 7m 20.74863290786743s) Loss: 0.00937(0.00937) Grad: 0.7944\nEpoch: [5][100/684] Data: 0.176 (0.195) Batch: 0.646 (0.664) Elapsed: 1m 7.096022367477417s (remain 6m 27.296841982567628s) Loss: 0.00107(0.00958) Grad: 0.0929\nEpoch: [5][200/684] Data: 0.173 (0.192) Batch: 0.642 (0.662) Elapsed: 2m 13.063384532928467s (remain 5m 19.749327011962407s) Loss: 0.01356(0.00961) Grad: 2.0370\nEpoch: [5][300/684] Data: 0.214 (0.192) Batch: 0.683 (0.662) Elapsed: 3m 19.212665796279907s (remain 4m 13.483225913538888s) Loss: 0.00062(0.00901) Grad: 0.1106\nEpoch: [5][400/684] Data: 0.194 (0.192) Batch: 0.663 (0.662) Elapsed: 4m 25.34626817703247s (remain 3m 7.264323925436884s) Loss: 0.02366(0.00904) Grad: 2.9149\nEpoch: [5][500/684] Data: 0.212 (0.191) Batch: 0.681 (0.661) Elapsed: 5m 31.284503698349s (remain 2m 1.0081121293371211s) Loss: 0.00077(0.00958) Grad: 0.0769\nEpoch: [5][600/684] Data: 0.173 (0.190) Batch: 0.643 (0.660) Elapsed: 6m 36.87773013114929s (remain 0m 54.81006921944328s) Loss: 0.00073(0.00994) Grad: 0.0649\nEpoch: [5][683/684] Data: 0.157 (0.190) Batch: 0.560 (0.660) Elapsed: 7m 31.12494683265686s (remain 0m 0.0s) Loss: 0.00385(0.01052) Grad: 0.6005\nEval: [0/171] Data: 0.174 (0.174) Batch: 0.314 (0.314) Elapsed: 0m 0.31400203704833984s (remain 0m 53.38034629821777s) Loss: 0.00358 (0.00358)\nEval: [100/171] Data: 0.182 (0.183) Batch: 0.321 (0.323) Elapsed: 0m 32.60361623764038s (remain 0m 22.596565709255707s) Loss: 0.00029 (0.19671)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 5 - avg_epoch_loss: 0.01052 - avg_val_loss: 0.19107 - time: 507s\n===== STARTING FOLD 3: ======\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.171 (0.184) Batch: 0.305 (0.324) Elapsed: 0m 55.34989666938782s (remain 0m 0.0s) Loss: 0.07290 (0.19107)\nEpoch: [1][0/684] Data: 0.174 (0.174) Batch: 0.645 (0.645) Elapsed: 0m 0.6448991298675537s (remain 7m 20.466105699539185s) Loss: 0.65557(0.65557) Grad: 3.3966\nEpoch: [1][100/684] Data: 0.184 (0.185) Batch: 0.653 (0.655) Elapsed: 1m 6.113469362258911s (remain 6m 21.625273645514312s) Loss: 0.09617(0.14757) Grad: 0.8408\nEpoch: [1][200/684] Data: 0.176 (0.187) Batch: 0.645 (0.657) Elapsed: 2m 12.011701583862305s (remain 5m 17.22214858211686s) Loss: 0.22769(0.12829) Grad: 2.1671\nEpoch: [1][300/684] Data: 0.179 (0.187) Batch: 0.649 (0.657) Elapsed: 3m 17.729042530059814s (remain 4m 11.595426209345192s) Loss: 0.07813(0.12025) Grad: 0.8066\nEpoch: [1][400/684] Data: 0.173 (0.187) Batch: 0.642 (0.657) Elapsed: 4m 23.312915802001953s (remain 3m 5.829314643308123s) Loss: 0.08187(0.11533) Grad: 0.7493\nEpoch: [1][500/684] Data: 0.176 (0.187) Batch: 0.647 (0.657) Elapsed: 5m 29.174299716949463s (remain 2m 0.23731905828697109s) Loss: 0.01974(0.11165) Grad: 0.2612\nEpoch: [1][600/684] Data: 0.177 (0.187) Batch: 0.646 (0.657) Elapsed: 6m 34.83259320259094s (remain 0m 54.527629344118225s) Loss: 0.02377(0.11099) Grad: 0.3273\nEpoch: [1][683/684] Data: 0.150 (0.187) Batch: 0.564 (0.656) Elapsed: 7m 28.93990707397461s (remain 0m 0.0s) Loss: 0.08542(0.10831) Grad: 0.6442\nEval: [0/171] Data: 0.169 (0.169) Batch: 0.309 (0.309) Elapsed: 0m 0.30905818939208984s (remain 0m 52.53989219665527s) Loss: 0.01802 (0.01802)\nEval: [100/171] Data: 0.173 (0.183) Batch: 0.313 (0.323) Elapsed: 0m 32.57724976539612s (remain 0m 22.57829191661117s) Loss: 0.01760 (0.09584)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 1 - avg_epoch_loss: 0.10831 - avg_val_loss: 0.10152 - time: 504s\nEpoch: 1 - Save best score: 0.0163\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.178 (0.185) Batch: 0.308 (0.324) Elapsed: 0m 55.46665096282959s (remain 0m 0.0s) Loss: 0.15401 (0.10152)\nEpoch: [2][0/684] Data: 0.172 (0.172) Batch: 0.641 (0.641) Elapsed: 0m 0.6410348415374756s (remain 7m 17.826796770095825s) Loss: 0.12507(0.12507) Grad: 0.7531\nEpoch: [2][100/684] Data: 0.171 (0.185) Batch: 0.641 (0.655) Elapsed: 1m 6.117372751235962s (remain 6m 21.64780508881745s) Loss: 0.08543(0.08051) Grad: 0.7896\nEpoch: [2][200/684] Data: 0.183 (0.190) Batch: 0.653 (0.660) Elapsed: 2m 12.697200775146484s (remain 5m 18.869392907441522s) Loss: 0.02053(0.08689) Grad: 0.2915\nEpoch: [2][300/684] Data: 0.179 (0.189) Batch: 0.648 (0.658) Elapsed: 3m 18.19034695625305s (remain 4m 12.182401608787075s) Loss: 0.19803(0.08686) Grad: 1.7185\nEpoch: [2][400/684] Data: 0.175 (0.188) Batch: 0.648 (0.657) Elapsed: 4m 23.601735830307007s (remain 3m 6.033145236850089s) Loss: 0.10084(0.08791) Grad: 1.4869\nEpoch: [2][500/684] Data: 0.172 (0.188) Batch: 0.641 (0.658) Elapsed: 5m 29.60198998451233s (remain 2m 0.3935412518279122s) Loss: 0.12574(0.08809) Grad: 1.1200\nEpoch: [2][600/684] Data: 0.190 (0.187) Batch: 0.659 (0.657) Elapsed: 6m 35.08302402496338s (remain 0m 54.562214632399275s) Loss: 0.06055(0.08886) Grad: 0.8118\nEpoch: [2][683/684] Data: 0.251 (0.187) Batch: 0.666 (0.657) Elapsed: 7m 29.46588444709778s (remain 0m 0.0s) Loss: 0.01523(0.08995) Grad: 0.2035\nEval: [0/171] Data: 0.206 (0.206) Batch: 0.345 (0.345) Elapsed: 0m 0.34496068954467773s (remain 0m 58.643317222595215s) Loss: 0.01160 (0.01160)\nEval: [100/171] Data: 0.190 (0.196) Batch: 0.330 (0.336) Elapsed: 0m 33.9329617023468s (remain 0m 23.517894249151247s) Loss: 0.01329 (0.09727)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 2 - avg_epoch_loss: 0.08995 - avg_val_loss: 0.10191 - time: 506s\nEpoch: 2 - Save best score: 0.0187\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.161 (0.191) Batch: 0.291 (0.330) Elapsed: 0m 56.45977306365967s (remain 0m 0.0s) Loss: 0.17922 (0.10191)\nEpoch: [3][0/684] Data: 0.189 (0.189) Batch: 0.659 (0.659) Elapsed: 0m 0.6590185165405273s (remain 7m 30.109646797180176s) Loss: 0.01664(0.01664) Grad: 0.2384\nEpoch: [3][100/684] Data: 0.173 (0.186) Batch: 0.642 (0.656) Elapsed: 1m 6.2339582443237305s (remain 6m 22.3207688756508s) Loss: 0.10350(0.06077) Grad: 3.1000\nEpoch: [3][200/684] Data: 0.178 (0.189) Batch: 0.647 (0.659) Elapsed: 2m 12.469874143600464s (remain 5m 18.323130404771234s) Loss: 0.27752(0.06137) Grad: 4.7991\nEpoch: [3][300/684] Data: 0.179 (0.190) Batch: 0.648 (0.660) Elapsed: 3m 18.619821071624756s (remain 4m 12.728875317050779s) Loss: 0.05223(0.06370) Grad: 1.9485\nEpoch: [3][400/684] Data: 0.198 (0.190) Batch: 0.667 (0.660) Elapsed: 4m 24.86158847808838s (remain 3m 6.922268177803005s) Loss: 0.03066(0.06384) Grad: 1.0993\nEpoch: [3][500/684] Data: 0.179 (0.189) Batch: 0.648 (0.659) Elapsed: 5m 30.404701709747314s (remain 2m 0.6867473311053232s) Loss: 0.11987(0.06490) Grad: 2.6697\nEpoch: [3][600/684] Data: 0.195 (0.189) Batch: 0.664 (0.659) Elapsed: 6m 36.08584213256836s (remain 0m 54.70070698336633s) Loss: 0.02547(0.06384) Grad: 0.5772\nEpoch: [3][683/684] Data: 0.163 (0.189) Batch: 0.577 (0.659) Elapsed: 7m 30.635140895843506s (remain 0m 0.0s) Loss: 0.02975(0.06317) Grad: 2.2498\nEval: [0/171] Data: 0.191 (0.191) Batch: 0.331 (0.331) Elapsed: 0m 0.330737829208374s (remain 0m 56.225430965423584s) Loss: 0.01286 (0.01286)\nEval: [100/171] Data: 0.248 (0.190) Batch: 0.388 (0.330) Elapsed: 0m 33.32913255691528s (remain 0m 23.099398801822474s) Loss: 0.01377 (0.11366)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 3 - avg_epoch_loss: 0.06317 - avg_val_loss: 0.12016 - time: 507s\nEpoch: 3 - Save best score: 0.0365\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.165 (0.188) Batch: 0.296 (0.328) Elapsed: 0m 56.09799361228943s (remain 0m 0.0s) Loss: 0.32625 (0.12016)\nEpoch: [4][0/684] Data: 0.182 (0.182) Batch: 0.650 (0.650) Elapsed: 0m 0.6502866744995117s (remain 7m 24.145798683166504s) Loss: 0.01077(0.01077) Grad: 0.2066\nEpoch: [4][100/684] Data: 0.201 (0.191) Batch: 0.670 (0.661) Elapsed: 1m 6.809443473815918s (remain 6m 25.642629160739375s) Loss: 0.05804(0.01909) Grad: 5.0962\nEpoch: [4][200/684] Data: 0.169 (0.188) Batch: 0.638 (0.658) Elapsed: 2m 12.288364171981812s (remain 5m 17.886964652075676s) Loss: 0.00245(0.01895) Grad: 0.2701\nEpoch: [4][300/684] Data: 0.237 (0.188) Batch: 0.711 (0.658) Elapsed: 3m 18.04075312614441s (remain 4m 11.992054642236894s) Loss: 0.00751(0.01863) Grad: 1.1696\nEpoch: [4][400/684] Data: 0.221 (0.188) Batch: 0.690 (0.658) Elapsed: 4m 23.761228799819946s (remain 3m 6.145705113089889s) Loss: 0.01306(0.02058) Grad: 1.3590\nEpoch: [4][500/684] Data: 0.192 (0.188) Batch: 0.662 (0.658) Elapsed: 5m 29.761878490447998s (remain 2m 0.45194364022353284s) Loss: 0.00725(0.02224) Grad: 0.6446\nEpoch: [4][600/684] Data: 0.178 (0.188) Batch: 0.647 (0.658) Elapsed: 6m 35.47023606300354s (remain 0m 54.615689838983826s) Loss: 0.02294(0.02374) Grad: 2.5558\nEpoch: [4][683/684] Data: 0.154 (0.188) Batch: 0.572 (0.658) Elapsed: 7m 29.771278619766235s (remain 0m 0.0s) Loss: 0.01671(0.02509) Grad: 1.2041\nEval: [0/171] Data: 0.222 (0.222) Batch: 0.362 (0.362) Elapsed: 0m 0.36237382888793945s (remain 1m 1.603550910949707s) Loss: 0.00653 (0.00653)\nEval: [100/171] Data: 0.188 (0.186) Batch: 0.328 (0.325) Elapsed: 0m 32.83655667304993s (remain 0m 22.758009575381138s) Loss: 0.01845 (0.14507)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 4 - avg_epoch_loss: 0.02509 - avg_val_loss: 0.14816 - time: 505s\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.169 (0.185) Batch: 0.300 (0.324) Elapsed: 0m 55.457990646362305s (remain 0m 0.0s) Loss: 0.26513 (0.14816)\nEpoch: [5][0/684] Data: 0.177 (0.177) Batch: 0.645 (0.645) Elapsed: 0m 0.6453063488006592s (remain 7m 20.74423623085022s) Loss: 0.01336(0.01336) Grad: 1.9619\nEpoch: [5][100/684] Data: 0.176 (0.188) Batch: 0.646 (0.658) Elapsed: 1m 6.440087795257568s (remain 6m 23.510605788466933s) Loss: 0.00272(0.00848) Grad: 0.2799\nEpoch: [5][200/684] Data: 0.189 (0.188) Batch: 0.659 (0.658) Elapsed: 2m 12.285539150238037s (remain 5m 17.880176166989884s) Loss: 0.00238(0.00853) Grad: 0.4550\nEpoch: [5][300/684] Data: 0.226 (0.188) Batch: 0.703 (0.658) Elapsed: 3m 18.18585205078125s (remain 4m 12.176682177572161s) Loss: 0.01717(0.00979) Grad: 2.4224\nEpoch: [5][400/684] Data: 0.188 (0.188) Batch: 0.657 (0.658) Elapsed: 4m 23.918184757232666s (remain 3m 6.25647452941854s) Loss: 0.14002(0.01029) Grad: 5.8317\nEpoch: [5][500/684] Data: 0.174 (0.188) Batch: 0.644 (0.658) Elapsed: 5m 29.49609684944153s (remain 2m 0.354861723448721s) Loss: 0.06057(0.01035) Grad: 5.4475\nEpoch: [5][600/684] Data: 0.172 (0.187) Batch: 0.641 (0.657) Elapsed: 6m 35.07513165473938s (remain 0m 54.56112467112041s) Loss: 0.06632(0.01097) Grad: 4.9152\nEpoch: [5][683/684] Data: 0.228 (0.187) Batch: 0.648 (0.657) Elapsed: 7m 29.216002225875854s (remain 0m 0.0s) Loss: 0.00354(0.01149) Grad: 0.6290\nEval: [0/171] Data: 0.223 (0.223) Batch: 0.363 (0.363) Elapsed: 0m 0.36272454261779785s (remain 1m 1.6631722450256348s) Loss: 0.00145 (0.00145)\nEval: [100/171] Data: 0.208 (0.189) Batch: 0.348 (0.329) Elapsed: 0m 33.18589973449707s (remain 0m 23.000128528859356s) Loss: 0.00315 (0.18453)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 5 - avg_epoch_loss: 0.01149 - avg_val_loss: 0.19421 - time: 505s\n===== STARTING FOLD 4: ======\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.163 (0.186) Batch: 0.294 (0.326) Elapsed: 0m 55.77518367767334s (remain 0m 0.0s) Loss: 0.35805 (0.19421)\nEpoch: [1][0/684] Data: 0.181 (0.181) Batch: 0.651 (0.651) Elapsed: 0m 0.6514153480529785s (remain 7m 24.916682720184326s) Loss: 0.72469(0.72469) Grad: 3.9442\nEpoch: [1][100/684] Data: 0.176 (0.187) Batch: 0.647 (0.658) Elapsed: 1m 6.412131071090698s (remain 6m 23.349231826196785s) Loss: 0.02240(0.15360) Grad: 0.3021\nEpoch: [1][200/684] Data: 0.170 (0.186) Batch: 0.639 (0.656) Elapsed: 2m 11.847162008285522s (remain 5m 16.826762437820378s) Loss: 0.13722(0.13146) Grad: 1.4120\nEpoch: [1][300/684] Data: 0.202 (0.186) Batch: 0.672 (0.656) Elapsed: 3m 17.483847618103027s (remain 4m 11.283434012403518s) Loss: 0.02330(0.12289) Grad: 0.2863\nEpoch: [1][400/684] Data: 0.202 (0.186) Batch: 0.674 (0.656) Elapsed: 4m 22.93452548980713s (remain 3m 5.562271106272874s) Loss: 0.08431(0.11784) Grad: 0.5539\nEpoch: [1][500/684] Data: 0.182 (0.187) Batch: 0.652 (0.657) Elapsed: 5m 29.200986862182617s (remain 2m 0.24706705744398505s) Loss: 0.13281(0.11525) Grad: 0.8037\nEpoch: [1][600/684] Data: 0.173 (0.187) Batch: 0.642 (0.657) Elapsed: 6m 35.01035690307617s (remain 0m 54.55217907313698s) Loss: 0.01797(0.11295) Grad: 0.2230\nEpoch: [1][683/684] Data: 0.207 (0.187) Batch: 0.624 (0.656) Elapsed: 7m 29.021808862686157s (remain 0m 0.0s) Loss: 0.02000(0.11002) Grad: 0.2613\nEval: [0/171] Data: 0.366 (0.366) Batch: 0.507 (0.507) Elapsed: 0m 0.5068187713623047s (remain 1m 26.159191131591797s) Loss: 0.14350 (0.14350)\nEval: [100/171] Data: 0.248 (0.185) Batch: 0.388 (0.325) Elapsed: 0m 32.81197500228882s (remain 0m 22.740972773863533s) Loss: 0.01919 (0.09912)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 1 - avg_epoch_loss: 0.11002 - avg_val_loss: 0.10070 - time: 504s\nEpoch: 1 - Save best score: 0.0184\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.183 (0.184) Batch: 0.318 (0.323) Elapsed: 0m 55.2689688205719s (remain 0m 0.0s) Loss: 0.03127 (0.10070)\nEpoch: [2][0/684] Data: 0.175 (0.175) Batch: 0.644 (0.644) Elapsed: 0m 0.6443166732788086s (remain 7m 20.06828784942627s) Loss: 0.02246(0.02246) Grad: 0.2584\nEpoch: [2][100/684] Data: 0.174 (0.186) Batch: 0.643 (0.656) Elapsed: 1m 6.268506050109863s (remain 6m 22.520188388257907s) Loss: 0.09365(0.08581) Grad: 0.6477\nEpoch: [2][200/684] Data: 0.188 (0.188) Batch: 0.657 (0.658) Elapsed: 2m 12.208855152130127s (remain 5m 17.695905664073848s) Loss: 0.02187(0.08556) Grad: 0.2768\nEpoch: [2][300/684] Data: 0.270 (0.187) Batch: 0.740 (0.657) Elapsed: 3m 17.70207405090332s (remain 4m 11.561110835534748s) Loss: 0.10648(0.08649) Grad: 1.9794\nEpoch: [2][400/684] Data: 0.177 (0.186) Batch: 0.646 (0.655) Elapsed: 4m 22.797635316848755s (remain 3m 5.465662829596511s) Loss: 0.07943(0.08580) Grad: 0.8680\nEpoch: [2][500/684] Data: 0.180 (0.185) Batch: 0.649 (0.655) Elapsed: 5m 28.110305309295654s (remain 1m 59.84867439441342s) Loss: 0.10735(0.08831) Grad: 0.8521\nEpoch: [2][600/684] Data: 0.202 (0.185) Batch: 0.670 (0.655) Elapsed: 6m 33.38412880897522s (remain 0m 54.32759183218792s) Loss: 0.22562(0.09001) Grad: 1.4792\nEpoch: [2][683/684] Data: 0.125 (0.185) Batch: 0.521 (0.654) Elapsed: 7m 27.523000478744507s (remain 0m 0.0s) Loss: 0.05018(0.08968) Grad: 0.8447\nEval: [0/171] Data: 0.167 (0.167) Batch: 0.306 (0.306) Elapsed: 0m 0.30599474906921387s (remain 0m 52.01910734176636s) Loss: 0.12811 (0.12811)\nEval: [100/171] Data: 0.221 (0.183) Batch: 0.362 (0.322) Elapsed: 0m 32.56936454772949s (remain 0m 22.572826914267964s) Loss: 0.02122 (0.09995)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 2 - avg_epoch_loss: 0.08968 - avg_val_loss: 0.10292 - time: 503s\nEpoch: 2 - Save best score: 0.0324\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.175 (0.182) Batch: 0.310 (0.321) Elapsed: 0m 54.947112798690796s (remain 0m 0.0s) Loss: 0.07225 (0.10292)\nEpoch: [3][0/684] Data: 0.173 (0.173) Batch: 0.642 (0.642) Elapsed: 0m 0.642279863357544s (remain 7m 18.677146673202515s) Loss: 0.03642(0.03642) Grad: 0.3786\nEpoch: [3][100/684] Data: 0.176 (0.189) Batch: 0.644 (0.659) Elapsed: 1m 6.52548885345459s (remain 6m 24.003564371921016s) Loss: 0.07977(0.06535) Grad: 1.5267\nEpoch: [3][200/684] Data: 0.188 (0.191) Batch: 0.657 (0.661) Elapsed: 2m 12.810301542282104s (remain 5m 19.14117236279725s) Loss: 0.10511(0.06255) Grad: 3.8523\nEpoch: [3][300/684] Data: 0.227 (0.191) Batch: 0.702 (0.660) Elapsed: 3m 18.791732788085938s (remain 4m 12.94762012570402s) Loss: 0.11319(0.06157) Grad: 2.4688\nEpoch: [3][400/684] Data: 0.171 (0.190) Batch: 0.640 (0.660) Elapsed: 4m 24.541032075881958s (remain 3m 6.696040093452837s) Loss: 0.01349(0.06203) Grad: 0.4606\nEpoch: [3][500/684] Data: 0.181 (0.189) Batch: 0.650 (0.659) Elapsed: 5m 30.130950450897217s (remain 2m 0.5867543563157938s) Loss: 0.15154(0.06243) Grad: 2.3678\nEpoch: [3][600/684] Data: 0.175 (0.189) Batch: 0.644 (0.659) Elapsed: 6m 35.8544545173645s (remain 0m 54.66875162219844s) Loss: 0.02332(0.06261) Grad: 0.7758\nEpoch: [3][683/684] Data: 0.127 (0.188) Batch: 0.524 (0.658) Elapsed: 7m 30.07879066467285s (remain 0m 0.0s) Loss: 0.01475(0.06427) Grad: 0.2944\nEval: [0/171] Data: 0.169 (0.169) Batch: 0.309 (0.309) Elapsed: 0m 0.30918383598327637s (remain 0m 52.56125211715698s) Loss: 0.14810 (0.14810)\nEval: [100/171] Data: 0.247 (0.183) Batch: 0.387 (0.323) Elapsed: 0m 32.60421538352966s (remain 0m 22.596980958881943s) Loss: 0.01206 (0.10726)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 3 - avg_epoch_loss: 0.06427 - avg_val_loss: 0.11021 - time: 505s\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.174 (0.183) Batch: 0.309 (0.323) Elapsed: 0m 55.17231249809265s (remain 0m 0.0s) Loss: 0.04910 (0.11021)\nEpoch: [4][0/684] Data: 0.216 (0.216) Batch: 0.686 (0.686) Elapsed: 0m 0.685666561126709s (remain 7m 48.310261249542236s) Loss: 0.03168(0.03168) Grad: 0.8072\nEpoch: [4][100/684] Data: 0.179 (0.189) Batch: 0.648 (0.659) Elapsed: 1m 6.579503774642944s (remain 6m 24.315353471453818s) Loss: 0.00385(0.02068) Grad: 0.2007\nEpoch: [4][200/684] Data: 0.172 (0.187) Batch: 0.641 (0.657) Elapsed: 2m 12.114110708236694s (remain 5m 17.468236179494113s) Loss: 0.00834(0.02302) Grad: 1.2898\nEpoch: [4][300/684] Data: 0.242 (0.187) Batch: 0.716 (0.658) Elapsed: 3m 17.913830041885376s (remain 4m 11.830554505123246s) Loss: 0.00240(0.02505) Grad: 0.1075\nEpoch: [4][400/684] Data: 0.177 (0.188) Batch: 0.646 (0.658) Elapsed: 4m 23.92817449569702s (remain 3m 6.263524644095412s) Loss: 0.01176(0.02516) Grad: 1.2849\nEpoch: [4][500/684] Data: 0.190 (0.188) Batch: 0.659 (0.658) Elapsed: 5m 29.839808225631714s (remain 2m 0.48040899259603975s) Loss: 0.00594(0.02560) Grad: 0.9586\nEpoch: [4][600/684] Data: 0.196 (0.188) Batch: 0.665 (0.658) Elapsed: 6m 35.38235569000244s (remain 0m 54.60355328164758s) Loss: 0.01224(0.02592) Grad: 1.3007\nEpoch: [4][683/684] Data: 0.125 (0.187) Batch: 0.523 (0.657) Elapsed: 7m 29.507813453674316s (remain 0m 0.0s) Loss: 0.03340(0.02599) Grad: 2.9180\nEval: [0/171] Data: 0.277 (0.277) Batch: 0.417 (0.417) Elapsed: 0m 0.41740989685058594s (remain 1m 10.95968246459961s) Loss: 0.06468 (0.06468)\nEval: [100/171] Data: 0.190 (0.190) Batch: 0.330 (0.330) Elapsed: 0m 33.332456827163696s (remain 0m 23.101702751499587s) Loss: 0.03381 (0.17656)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 4 - avg_epoch_loss: 0.02599 - avg_val_loss: 0.17817 - time: 505s\nEpoch: 4 - Save best score: 0.0643\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.171 (0.187) Batch: 0.306 (0.327) Elapsed: 0m 55.88433313369751s (remain 0m 0.0s) Loss: 0.19754 (0.17817)\nEpoch: [5][0/684] Data: 0.174 (0.174) Batch: 0.644 (0.644) Elapsed: 0m 0.6440954208374023s (remain 7m 19.9171724319458s) Loss: 0.00758(0.00758) Grad: 0.5888\nEpoch: [5][100/684] Data: 0.174 (0.186) Batch: 0.644 (0.656) Elapsed: 1m 6.227049112319946s (remain 6m 22.28088745032204s) Loss: 0.00405(0.01096) Grad: 0.5498\nEpoch: [5][200/684] Data: 0.179 (0.186) Batch: 0.649 (0.656) Elapsed: 2m 11.844696283340454s (remain 5m 16.820837337579235s) Loss: 0.00107(0.00947) Grad: 0.0732\nEpoch: [5][300/684] Data: 0.228 (0.187) Batch: 0.697 (0.657) Elapsed: 3m 17.79385209083557s (remain 4m 11.67789153086386s) Loss: 0.02704(0.00853) Grad: 4.2431\nEpoch: [5][400/684] Data: 0.173 (0.187) Batch: 0.643 (0.657) Elapsed: 4m 23.26184320449829s (remain 3m 5.7932708899576255s) Loss: 0.00547(0.00937) Grad: 1.3134\nEpoch: [5][500/684] Data: 0.176 (0.187) Batch: 0.646 (0.657) Elapsed: 5m 29.299113512039185s (remain 2m 0.2829097259544824s) Loss: 0.00202(0.00981) Grad: 0.2215\nEpoch: [5][600/684] Data: 0.187 (0.187) Batch: 0.656 (0.657) Elapsed: 6m 34.84735941886902s (remain 0m 54.529668605268114s) Loss: 0.00163(0.00991) Grad: 0.2388\nEpoch: [5][683/684] Data: 0.175 (0.187) Batch: 0.577 (0.657) Elapsed: 7m 29.192466974258423s (remain 0m 0.0s) Loss: 0.01355(0.01047) Grad: 2.1366\nEval: [0/171] Data: 0.235 (0.235) Batch: 0.375 (0.375) Elapsed: 0m 0.3748905658721924s (remain 1m 3.731396198272705s) Loss: 0.41933 (0.41933)\nEval: [100/171] Data: 0.175 (0.188) Batch: 0.314 (0.328) Elapsed: 0m 33.10963177680969s (remain 0m 22.94726954828394s) Loss: 0.13780 (0.33192)\n","output_type":"stream"},{"name":"stderr","text":"Epoch: 5 - avg_epoch_loss: 0.01047 - avg_val_loss: 0.34115 - time: 505s\nEpoch: 5 - Save best score: 0.0653\n","output_type":"stream"},{"name":"stdout","text":"Eval: [170/171] Data: 0.170 (0.187) Batch: 0.306 (0.327) Elapsed: 0m 55.864800691604614s (remain 0m 0.0s) Loss: 0.46384 (0.34115)\n","output_type":"stream"}]},{"cell_type":"markdown","source":"# CV Score","metadata":{}},{"cell_type":"code","source":"cv_score = pfbeta(oof_df.labels.values, oof_df.preds.values, beta=1)\nprint(f\"CV Score: {cv_score:.5f}\")","metadata":{"execution":{"iopub.status.busy":"2023-03-14T05:36:50.751482Z","iopub.execute_input":"2023-03-14T05:36:50.751842Z","iopub.status.idle":"2023-03-14T05:36:50.973824Z","shell.execute_reply.started":"2023-03-14T05:36:50.751807Z","shell.execute_reply":"2023-03-14T05:36:50.972862Z"},"trusted":true},"execution_count":23,"outputs":[{"name":"stdout","text":"CV Score: 0.04911\n","output_type":"stream"}]},{"cell_type":"code","source":"oof_df.labels","metadata":{"execution":{"iopub.status.busy":"2023-03-14T06:03:39.513764Z","iopub.execute_input":"2023-03-14T06:03:39.514462Z","iopub.status.idle":"2023-03-14T06:03:39.523065Z","shell.execute_reply.started":"2023-03-14T06:03:39.514417Z","shell.execute_reply":"2023-03-14T06:03:39.521794Z"},"trusted":true},"execution_count":29,"outputs":[{"execution_count":29,"output_type":"execute_result","data":{"text/plain":"0        0\n1        0\n2        0\n3        0\n4        0\n        ..\n10937    0\n10938    0\n10939    0\n10940    0\n10941    0\nName: labels, Length: 54706, dtype: int64"},"metadata":{}}]},{"cell_type":"code","source":"from sklearn.metrics import accuracy_score\n\n# print(Y_preds[:15])\n# print(Y_test[:15])\n\nprint('Test Accuracy     : {:.3f}'.format(accuracy_score(oof_df['labels'].to_list(), oof_df['preds'])))","metadata":{"execution":{"iopub.status.busy":"2023-03-14T06:06:19.772619Z","iopub.execute_input":"2023-03-14T06:06:19.772997Z","iopub.status.idle":"2023-03-14T06:06:19.800031Z","shell.execute_reply.started":"2023-03-14T06:06:19.772958Z","shell.execute_reply":"2023-03-14T06:06:19.798587Z"},"trusted":true},"execution_count":34,"outputs":[{"traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)","\u001b[0;32m/tmp/ipykernel_24/3780263671.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;31m# print(Y_test[:15])\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'Test Accuracy     : {:.3f}'\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0maccuracy_score\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moof_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'labels'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moof_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'preds'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m","\u001b[0;32m/opt/conda/lib/python3.7/site-packages/sklearn/metrics/_classification.py\u001b[0m in \u001b[0;36maccuracy_score\u001b[0;34m(y_true, y_pred, normalize, sample_weight)\u001b[0m\n\u001b[1;32m    209\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    210\u001b[0m     \u001b[0;31m# Compute accuracy for each possible representation\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 211\u001b[0;31m     \u001b[0my_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_true\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_pred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_check_targets\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_true\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_pred\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    212\u001b[0m     \u001b[0mcheck_consistent_length\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_true\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_pred\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msample_weight\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    213\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0my_type\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstartswith\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"multilabel\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.7/site-packages/sklearn/metrics/_classification.py\u001b[0m in \u001b[0;36m_check_targets\u001b[0;34m(y_true, y_pred)\u001b[0m\n\u001b[1;32m     93\u001b[0m         raise ValueError(\n\u001b[1;32m     94\u001b[0m             \"Classification metrics can't handle a mix of {0} and {1} targets\".format(\n\u001b[0;32m---> 95\u001b[0;31m                 \u001b[0mtype_true\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtype_pred\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     96\u001b[0m             )\n\u001b[1;32m     97\u001b[0m         )\n","\u001b[0;31mValueError\u001b[0m: Classification metrics can't handle a mix of binary and continuous targets"],"ename":"ValueError","evalue":"Classification metrics can't handle a mix of binary and continuous targets","output_type":"error"}]},{"cell_type":"code","source":"oof_df['preds'] = pd.to_numeric(oof_df['preds'], downcast='integer')\noof_df","metadata":{"execution":{"iopub.status.busy":"2023-03-14T06:15:37.816138Z","iopub.execute_input":"2023-03-14T06:15:37.817102Z","iopub.status.idle":"2023-03-14T06:15:37.840843Z","shell.execute_reply.started":"2023-03-14T06:15:37.817065Z","shell.execute_reply":"2023-03-14T06:15:37.83987Z"},"trusted":true},"execution_count":46,"outputs":[{"execution_count":46,"output_type":"execute_result","data":{"text/plain":"       labels     preds  fold  result\n0           0  0.000115     0   False\n1           0  0.008437     0   False\n2           0  0.000451     0   False\n3           0  0.031036     0   False\n4           0  0.009614     0   False\n...       ...       ...   ...     ...\n10937       0  0.047902     4   False\n10938       0  0.000009     4   False\n10939       0  0.000021     4   False\n10940       0  0.000003     4   False\n10941       0  0.000114     4   False\n\n[54706 rows x 4 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>labels</th>\n      <th>preds</th>\n      <th>fold</th>\n      <th>result</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>0.000115</td>\n      <td>0</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0</td>\n      <td>0.008437</td>\n      <td>0</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0</td>\n      <td>0.000451</td>\n      <td>0</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0</td>\n      <td>0.031036</td>\n      <td>0</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0</td>\n      <td>0.009614</td>\n      <td>0</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>10937</th>\n      <td>0</td>\n      <td>0.047902</td>\n      <td>4</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>10938</th>\n      <td>0</td>\n      <td>0.000009</td>\n      <td>4</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>10939</th>\n      <td>0</td>\n      <td>0.000021</td>\n      <td>4</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>10940</th>\n      <td>0</td>\n      <td>0.000003</td>\n      <td>4</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>10941</th>\n      <td>0</td>\n      <td>0.000114</td>\n      <td>4</td>\n      <td>False</td>\n    </tr>\n  </tbody>\n</table>\n<p>54706 rows × 4 columns</p>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"preds = oof_df['preds'].to_list()\npreds = [round(num) for num in preds]\npreds","metadata":{"execution":{"iopub.status.busy":"2023-03-14T06:17:27.146019Z","iopub.execute_input":"2023-03-14T06:17:27.14671Z","iopub.status.idle":"2023-03-14T06:17:27.177627Z","shell.execute_reply.started":"2023-03-14T06:17:27.146672Z","shell.execute_reply":"2023-03-14T06:17:27.176563Z"},"trusted":true},"execution_count":48,"outputs":[{"execution_count":48,"output_type":"execute_result","data":{"text/plain":"[0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n ...]"},"metadata":{}}]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# oof_df['result'] = oof_df['labels'] == int(oof_df['preds'])\n# oof_df[oof_df['labels'] != oof_df['preds']]['result'] = False\noof_df","metadata":{"execution":{"iopub.status.busy":"2023-03-14T06:13:19.068074Z","iopub.execute_input":"2023-03-14T06:13:19.068475Z","iopub.status.idle":"2023-03-14T06:13:19.083279Z","shell.execute_reply.started":"2023-03-14T06:13:19.068442Z","shell.execute_reply":"2023-03-14T06:13:19.082247Z"},"trusted":true},"execution_count":44,"outputs":[{"execution_count":44,"output_type":"execute_result","data":{"text/plain":"       labels     preds  fold  result\n0           0  0.000115     0   False\n1           0  0.008437     0   False\n2           0  0.000451     0   False\n3           0  0.031036     0   False\n4           0  0.009614     0   False\n...       ...       ...   ...     ...\n10937       0  0.047902     4   False\n10938       0  0.000009     4   False\n10939       0  0.000021     4   False\n10940       0  0.000003     4   False\n10941       0  0.000114     4   False\n\n[54706 rows x 4 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>labels</th>\n      <th>preds</th>\n      <th>fold</th>\n      <th>result</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>0.000115</td>\n      <td>0</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0</td>\n      <td>0.008437</td>\n      <td>0</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0</td>\n      <td>0.000451</td>\n      <td>0</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0</td>\n      <td>0.031036</td>\n      <td>0</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0</td>\n      <td>0.009614</td>\n      <td>0</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>10937</th>\n      <td>0</td>\n      <td>0.047902</td>\n      <td>4</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>10938</th>\n      <td>0</td>\n      <td>0.000009</td>\n      <td>4</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>10939</th>\n      <td>0</td>\n      <td>0.000021</td>\n      <td>4</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>10940</th>\n      <td>0</td>\n      <td>0.000003</td>\n      <td>4</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>10941</th>\n      <td>0</td>\n      <td>0.000114</td>\n      <td>4</td>\n      <td>False</td>\n    </tr>\n  </tbody>\n</table>\n<p>54706 rows × 4 columns</p>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"from sklearn.metrics import confusion_matrix\n\nconf_mat = confusion_matrix(oof_df['labels'].to_list(), preds'].to_list())\nprint(conf_mat)","metadata":{"execution":{"iopub.status.busy":"2023-03-14T06:09:06.083799Z","iopub.execute_input":"2023-03-14T06:09:06.084154Z","iopub.status.idle":"2023-03-14T06:09:06.124266Z","shell.execute_reply.started":"2023-03-14T06:09:06.084123Z","shell.execute_reply":"2023-03-14T06:09:06.122895Z"},"trusted":true},"execution_count":38,"outputs":[{"traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)","\u001b[0;32m/tmp/ipykernel_24/472774096.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0msklearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmetrics\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mconfusion_matrix\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mconf_mat\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mconfusion_matrix\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moof_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'labels'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moof_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'preds'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      4\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconf_mat\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.7/site-packages/sklearn/metrics/_classification.py\u001b[0m in \u001b[0;36mconfusion_matrix\u001b[0;34m(y_true, y_pred, labels, sample_weight, normalize)\u001b[0m\n\u001b[1;32m    305\u001b[0m     \u001b[0;34m(\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    306\u001b[0m     \"\"\"\n\u001b[0;32m--> 307\u001b[0;31m     \u001b[0my_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_true\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_pred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_check_targets\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_true\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_pred\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    308\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0my_type\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32min\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0;34m\"binary\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"multiclass\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    309\u001b[0m         \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"%s is not supported\"\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0my_type\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.7/site-packages/sklearn/metrics/_classification.py\u001b[0m in \u001b[0;36m_check_targets\u001b[0;34m(y_true, y_pred)\u001b[0m\n\u001b[1;32m     93\u001b[0m         raise ValueError(\n\u001b[1;32m     94\u001b[0m             \"Classification metrics can't handle a mix of {0} and {1} targets\".format(\n\u001b[0;32m---> 95\u001b[0;31m                 \u001b[0mtype_true\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtype_pred\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     96\u001b[0m             )\n\u001b[1;32m     97\u001b[0m         )\n","\u001b[0;31mValueError\u001b[0m: Classification metrics can't handle a mix of binary and continuous targets"],"ename":"ValueError","evalue":"Classification metrics can't handle a mix of binary and continuous targets","output_type":"error"}]},{"cell_type":"code","source":"oof_df['labels'].to_list()","metadata":{"execution":{"iopub.status.busy":"2023-03-14T06:08:55.094011Z","iopub.execute_input":"2023-03-14T06:08:55.094403Z","iopub.status.idle":"2023-03-14T06:08:55.116835Z","shell.execute_reply.started":"2023-03-14T06:08:55.094369Z","shell.execute_reply":"2023-03-14T06:08:55.115895Z"},"trusted":true},"execution_count":37,"outputs":[{"execution_count":37,"output_type":"execute_result","data":{"text/plain":"[0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 1,\n 1,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 1,\n 1,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 1,\n 1,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 1,\n 1,\n 1,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n 0,\n ...]"},"metadata":{}}]},{"cell_type":"code","source":"from sklearn.metrics import mean_absolute_error,r2_score,mean_squared_error\nimport numpy as np\ndef run_experiment(model):\n    model.fit(X_train, y_train)\n    y_pred = model.predict(X_test)\n    print(\"R^2 : \", r2_score(y_test, y_pred))\n    print(\"MAE :\", mean_absolute_error(y_test,y_pred))\n    print(\"RMSE:\",np.sqrt(mean_squared_error(y_test, y_pred)))\n\nplot_confusion_matrix(model, X_test, y_test, cmap='GnBu')\n    plt.show()\n    print('Precision: %.3f' % precision_score(y_test, y_pred))\n    print('Recall: %.3f' % recall_score(y_test, y_pred))\n    print('F1: %.3f' % f1_score(y_test, y_pred))\n    print('Accuracy: %.3f' % accuracy_score(y_test, y_pred))","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"oof_df.head(20)","metadata":{"execution":{"iopub.status.busy":"2023-03-14T05:36:50.975479Z","iopub.execute_input":"2023-03-14T05:36:50.975853Z","iopub.status.idle":"2023-03-14T05:36:50.990969Z","shell.execute_reply.started":"2023-03-14T05:36:50.975816Z","shell.execute_reply":"2023-03-14T05:36:50.989928Z"},"trusted":true},"execution_count":24,"outputs":[{"execution_count":24,"output_type":"execute_result","data":{"text/plain":"    labels     preds  fold\n0        0  0.000115     0\n1        0  0.008437     0\n2        0  0.000451     0\n3        0  0.031036     0\n4        0  0.009614     0\n5        0  0.006772     0\n6        0  0.000038     0\n7        0  0.028135     0\n8        0  0.149199     0\n9        0  0.049131     0\n10       0  0.000171     0\n11       0  0.007707     0\n12       0  0.002353     0\n13       0  0.008714     0\n14       0  0.003949     0\n15       0  0.000946     0\n16       0  0.000775     0\n17       0  0.001270     0\n18       0  0.065220     0\n19       0  0.025244     0","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>labels</th>\n      <th>preds</th>\n      <th>fold</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>0.000115</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0</td>\n      <td>0.008437</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0</td>\n      <td>0.000451</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0</td>\n      <td>0.031036</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0</td>\n      <td>0.009614</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>0</td>\n      <td>0.006772</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>0</td>\n      <td>0.000038</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>0</td>\n      <td>0.028135</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>8</th>\n      <td>0</td>\n      <td>0.149199</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>9</th>\n      <td>0</td>\n      <td>0.049131</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>10</th>\n      <td>0</td>\n      <td>0.000171</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>11</th>\n      <td>0</td>\n      <td>0.007707</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td>0</td>\n      <td>0.002353</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>13</th>\n      <td>0</td>\n      <td>0.008714</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>14</th>\n      <td>0</td>\n      <td>0.003949</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>15</th>\n      <td>0</td>\n      <td>0.000946</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>16</th>\n      <td>0</td>\n      <td>0.000775</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>17</th>\n      <td>0</td>\n      <td>0.001270</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>18</th>\n      <td>0</td>\n      <td>0.065220</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>19</th>\n      <td>0</td>\n      <td>0.025244</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]}]}