Edit model card

Metrics

  • loss: 1.7243
  • accuracy: 0.7996
  • precision: 0.7969
  • recall: 0.7996
  • precision_macro: 0.6535
  • recall_macro: 0.6526
  • macro_fpr: 0.0942
  • weighted_fpr: 0.0771
  • weighted_specificity: 0.8638
  • macro_specificity: 0.9158
  • weighted_sensitivity: 0.7996
  • macro_sensitivity: 0.6526
  • f1_micro: 0.7996
  • f1_macro: 0.6529
  • f1_weighted: 0.7982
  • runtime: 351.9249
  • samples_per_second: 1.2760
  • steps_per_second: 0.1620

case-analysis-bert-base-uncased

This model is a fine-tuned version of google-bert/bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.7243
  • Accuracy: 0.7996
  • Precision: 0.7969
  • Recall: 0.7996
  • Precision Macro: 0.6427
  • Recall Macro: 0.6184
  • Macro Fpr: 0.0946
  • Weighted Fpr: 0.0712
  • Weighted Specificity: 0.8449
  • Macro Specificity: 0.9145
  • Weighted Sensitivity: 0.8129
  • Macro Sensitivity: 0.6184
  • F1 Micro: 0.8129
  • F1 Macro: 0.6284
  • F1 Weighted: 0.8035

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 30
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision Recall Precision Macro Recall Macro Macro Fpr Weighted Fpr Weighted Specificity Macro Specificity Weighted Sensitivity Macro Sensitivity F1 Micro F1 Macro F1 Weighted
No log 1.0 224 0.7283 0.7862 0.7487 0.7862 0.5848 0.5572 0.1142 0.0831 0.8036 0.8974 0.7862 0.5572 0.7862 0.5606 0.7597
No log 2.0 448 0.8160 0.7996 0.7603 0.7996 0.5770 0.6065 0.0997 0.0771 0.8417 0.9103 0.7996 0.6065 0.7996 0.5914 0.7794
0.6512 3.0 672 0.8588 0.7906 0.7598 0.7906 0.5770 0.5989 0.1005 0.0811 0.8512 0.9105 0.7906 0.5989 0.7906 0.5840 0.7720
0.6512 4.0 896 1.0821 0.7817 0.7819 0.7817 0.6214 0.6429 0.0996 0.0851 0.8679 0.9124 0.7817 0.6429 0.7817 0.6299 0.7805
0.3466 5.0 1120 1.0612 0.8085 0.7999 0.8085 0.7129 0.6263 0.0948 0.0732 0.8470 0.9139 0.8085 0.6263 0.8085 0.6195 0.7928
0.3466 6.0 1344 1.2559 0.7929 0.7877 0.7929 0.6206 0.6362 0.0951 0.0801 0.8717 0.9161 0.7929 0.6362 0.7929 0.6273 0.7897
0.1715 7.0 1568 1.3701 0.7929 0.7889 0.7929 0.6345 0.6179 0.0991 0.0801 0.8558 0.9122 0.7929 0.6179 0.7929 0.6237 0.7893
0.1715 8.0 1792 1.4005 0.8107 0.8035 0.8107 0.6578 0.6370 0.0922 0.0722 0.8607 0.9179 0.8107 0.6370 0.8107 0.6464 0.8064
0.0636 9.0 2016 1.4737 0.8018 0.7881 0.8018 0.6583 0.6149 0.1026 0.0761 0.8271 0.9072 0.8018 0.6149 0.8018 0.6263 0.7896
0.0636 10.0 2240 1.7569 0.7884 0.7962 0.7884 0.6275 0.6428 0.0960 0.0821 0.8750 0.9158 0.7884 0.6428 0.7884 0.6332 0.7909
0.0636 11.0 2464 1.7141 0.7906 0.7824 0.7906 0.6166 0.6083 0.1035 0.0811 0.8424 0.9083 0.7906 0.6083 0.7906 0.6101 0.7845
0.0159 12.0 2688 1.7144 0.7951 0.7914 0.7951 0.6393 0.6413 0.0969 0.0791 0.8610 0.9140 0.7951 0.6413 0.7951 0.6373 0.7917
0.0159 13.0 2912 1.7243 0.7996 0.7969 0.7996 0.6535 0.6526 0.0942 0.0771 0.8638 0.9158 0.7996 0.6526 0.7996 0.6529 0.7982
0.0043 14.0 3136 1.8551 0.7973 0.7948 0.7973 0.6576 0.6189 0.1041 0.0781 0.8314 0.9072 0.7973 0.6189 0.7973 0.6335 0.7912
0.0043 15.0 3360 1.8841 0.7929 0.7869 0.7929 0.6154 0.6162 0.1008 0.0801 0.8511 0.9110 0.7929 0.6162 0.7929 0.6104 0.7861
0.0029 16.0 3584 2.0853 0.7550 0.7837 0.7550 0.6010 0.6119 0.1100 0.0976 0.8698 0.9062 0.7550 0.6119 0.7550 0.6015 0.7661
0.0029 17.0 3808 1.9722 0.7840 0.7783 0.7840 0.6018 0.5839 0.1076 0.0841 0.8394 0.9059 0.7840 0.5839 0.7840 0.5917 0.7797
0.0071 18.0 4032 1.8735 0.7996 0.7783 0.7996 0.6086 0.5917 0.1053 0.0771 0.8193 0.9047 0.7996 0.5917 0.7996 0.5960 0.7840
0.0071 19.0 4256 1.8294 0.8018 0.7840 0.8018 0.6114 0.5943 0.1025 0.0761 0.8308 0.9082 0.8018 0.5943 0.8018 0.6001 0.7895
0.0071 20.0 4480 1.8578 0.7973 0.7939 0.7973 0.6367 0.6232 0.0990 0.0781 0.8497 0.9118 0.7973 0.6232 0.7973 0.6285 0.7942
0.0049 21.0 4704 1.8770 0.7973 0.7939 0.7973 0.6367 0.6232 0.0990 0.0781 0.8497 0.9118 0.7973 0.6232 0.7973 0.6285 0.7942
0.0049 22.0 4928 1.8932 0.7951 0.7876 0.7951 0.6219 0.6119 0.1007 0.0791 0.8461 0.9103 0.7951 0.6119 0.7951 0.6155 0.7900
0.0015 23.0 5152 1.9834 0.7996 0.7965 0.7996 0.6441 0.6389 0.0960 0.0771 0.8599 0.9149 0.7996 0.6389 0.7996 0.6403 0.7971
0.0015 24.0 5376 1.9926 0.8018 0.7984 0.8018 0.6468 0.6399 0.0952 0.0761 0.8603 0.9155 0.8018 0.6399 0.8018 0.6422 0.7991
0.0001 25.0 5600 1.9771 0.7973 0.7790 0.7973 0.6025 0.6024 0.1011 0.0781 0.8420 0.9098 0.7973 0.6024 0.7973 0.6017 0.7871
0.0001 26.0 5824 1.9871 0.7951 0.7770 0.7951 0.5996 0.6015 0.1020 0.0791 0.8416 0.9092 0.7951 0.6015 0.7951 0.5997 0.7850
0.0 27.0 6048 1.8756 0.8129 0.7961 0.8129 0.6440 0.6200 0.0939 0.0712 0.8462 0.9148 0.8129 0.6200 0.8129 0.6293 0.8029
0.0 28.0 6272 1.8473 0.8151 0.7998 0.8151 0.6463 0.6194 0.0937 0.0703 0.8453 0.9151 0.8151 0.6194 0.8151 0.6305 0.8056
0.0 29.0 6496 1.8525 0.8129 0.7975 0.8129 0.6427 0.6184 0.0946 0.0712 0.8449 0.9145 0.8129 0.6184 0.8129 0.6284 0.8035
0.0001 30.0 6720 1.8540 0.8129 0.7975 0.8129 0.6427 0.6184 0.0946 0.0712 0.8449 0.9145 0.8129 0.6184 0.8129 0.6284 0.8035

Framework versions

  • Transformers 4.40.1
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
18
Safetensors
Model size
109M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for cite-text-analysis/case-analysis-bert-base-uncased

Finetuned
(2133)
this model