File size: 2,547 Bytes
62bcbed f5eafb5 1c497fa c31e608 c1131e0 89dd885 8302160 41572be ffbcc89 cbdcaed 6e8c3e3 878748f fef6858 7a4c868 e65c35d c07b375 617b668 0e3d5da 0a11735 175e9cc c0c9574 2f7c918 bb2d485 75608ae d89731f 4a03687 df06d69 d57b0b2 7f4b8e4 2ce6446 064f70f 30fa70d 7a92f2e 1a00b6d 18326d2 39da080 756bee5 514a337 2bea2ff 084e2c8 7f2884a 77962e4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 |
Started at: 16:03:31
nb-bert-base, 1e-06, 128
({'_name_or_path': '/disk4/folder1/working/checkpoints/huggingface/native_pytorch/step4_8/', 'attention_probs_dropout_prob': 0.1, 'directionality': 'bidi', 'gradient_checkpointing': False, 'hidden_act': 'gelu', 'hidden_dropout_prob': 0.1, 'hidden_size': 768, 'initializer_range': 0.02, 'intermediate_size': 3072, 'layer_norm_eps': 1e-12, 'max_position_embeddings': 512, 'model_type': 'bert', 'num_attention_heads': 12, 'num_hidden_layers': 12, 'pad_token_id': 0, 'pooler_fc_size': 768, 'pooler_num_attention_heads': 12, 'pooler_num_fc_layers': 3, 'pooler_size_per_head': 128, 'pooler_type': 'first_token_transform', 'position_embedding_type': 'absolute', 'type_vocab_size': 2, 'vocab_size': 119547, '_commit_hash': '82b194c0b3ea1fcad65f1eceee04adb26f9f71ac'}, {})
Epoch: 0
Training loss: 0.7271318706599149 - MAE: 0.6831216644315611
Validation loss : 0.42457069094116623 - MAE: 0.5310234954937942
Epoch: 1
Training loss: 0.4017618325623599 - MAE: 0.5113891617478459
Validation loss : 0.3601061556790326 - MAE: 0.47872181276206954
Epoch: 2
Training loss: 0.3346875086426735 - MAE: 0.4557084281637824
Validation loss : 0.3026893593975016 - MAE: 0.42968294420152836
Epoch: 3
Training loss: 0.27502210925925863 - MAE: 0.4083413682966922
Validation loss : 0.2493422031402588 - MAE: 0.38829624389412676
Epoch: 4
Training loss: 0.22922010150822728 - MAE: 0.374115376863141
Validation loss : 0.21512102678015427 - MAE: 0.36273443011280543
Epoch: 5
Training loss: 0.20582241117954253 - MAE: 0.35785392938520194
Validation loss : 0.19260533154010773 - MAE: 0.3445584427568667
Epoch: 6
Training loss: 0.18504355576905337 - MAE: 0.33999207572682016
Validation loss : 0.18239208814260122 - MAE: 0.3362534719407864
Epoch: 7
Training loss: 0.17347032624212178 - MAE: 0.32901041370404
Validation loss : 0.1727527996172776 - MAE: 0.3273999893030842
Epoch: 8
Training loss: 0.16140718588774855 - MAE: 0.3185892364257231
Validation loss : 0.16311134518803777 - MAE: 0.31766631226614
Epoch: 9
Training loss: 0.1555855579674244 - MAE: 0.3131268902757298
Validation loss : 0.15599115515077436 - MAE: 0.31016476266703236
Epoch: 10
Training loss: 0.14619598415764895 - MAE: 0.303503172138375
Validation loss : 0.150120818937147 - MAE: 0.3037910577986695
Epoch: 11
Training loss: 0.14181994023648176 - MAE: 0.2996883375884051
Validation loss : 0.1466762089246028 - MAE: 0.2993610834634052
Epoch: 12
Training loss: 0.13435588756745512 - MAE: 0.29110208978373203
Validation loss : 0.14056996175566236 - MAE: 0.29225523563049816
|