update files
Browse files
README.md
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext_pub_section
|
| 2 |
+
- original model file name: textclassifer_BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext_pubmed_20k
|
| 3 |
+
- This is a fine-tuned checkpoint of `microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext` for document section text classification
|
| 4 |
+
- possible document section classes are:BACKGROUND, CONCLUSIONS, METHODS, OBJECTIVE, RESULTS,
|
| 5 |
+
|
| 6 |
+
## metadata
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
### training_metrics
|
| 10 |
+
|
| 11 |
+
- val_accuracy: 0.8724678754806519
|
| 12 |
+
|
| 13 |
+
- val_matthewscorrcoef: 0.8290084004402161
|
| 14 |
+
|
| 15 |
+
- val_f1score: 0.8707401752471924
|
| 16 |
+
|
| 17 |
+
- val_cross_entropy: 0.3521413803100586
|
| 18 |
+
|
| 19 |
+
- epoch: 4.0
|
| 20 |
+
|
| 21 |
+
- train_accuracy_step: 0.6875
|
| 22 |
+
|
| 23 |
+
- train_matthewscorrcoef_step: 0.5752688050270081
|
| 24 |
+
|
| 25 |
+
- train_f1score_step: 0.6875
|
| 26 |
+
|
| 27 |
+
- train_cross_entropy_step: 0.9718704223632812
|
| 28 |
+
|
| 29 |
+
- train_accuracy_epoch: 0.8582640886306763
|
| 30 |
+
|
| 31 |
+
- train_matthewscorrcoef_epoch: 0.8091862797737122
|
| 32 |
+
|
| 33 |
+
- train_f1score_epoch: 0.8565496206283569
|
| 34 |
+
|
| 35 |
+
- train_cross_entropy_epoch: 0.39623069763183594
|
| 36 |
+
|
| 37 |
+
- test_accuracy: 0.863082766532898
|
| 38 |
+
|
| 39 |
+
- test_matthewscorrcoef: 0.816676676273346
|
| 40 |
+
|
| 41 |
+
- test_f1score: 0.8613550662994385
|
| 42 |
+
|
| 43 |
+
- test_cross_entropy: 0.3769755959510803
|
| 44 |
+
|
| 45 |
+
- date_run: Apr-21-2022_t-02
|
| 46 |
+
|
| 47 |
+
- huggingface_tag: microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext
|
| 48 |
+
|
| 49 |
+
### training_parameters
|
| 50 |
+
|
| 51 |
+
- val_accuracy: 0.8724678754806519
|
| 52 |
+
|
| 53 |
+
- val_matthewscorrcoef: 0.8290084004402161
|
| 54 |
+
|
| 55 |
+
- val_f1score: 0.8707401752471924
|
| 56 |
+
|
| 57 |
+
- val_cross_entropy: 0.3521413803100586
|
| 58 |
+
|
| 59 |
+
- epoch: 4.0
|
| 60 |
+
|
| 61 |
+
- train_accuracy_step: 0.6875
|
| 62 |
+
|
| 63 |
+
- train_matthewscorrcoef_step: 0.5752688050270081
|
| 64 |
+
|
| 65 |
+
- train_f1score_step: 0.6875
|
| 66 |
+
|
| 67 |
+
- train_cross_entropy_step: 0.9718704223632812
|
| 68 |
+
|
| 69 |
+
- train_accuracy_epoch: 0.8582640886306763
|
| 70 |
+
|
| 71 |
+
- train_matthewscorrcoef_epoch: 0.8091862797737122
|
| 72 |
+
|
| 73 |
+
- train_f1score_epoch: 0.8565496206283569
|
| 74 |
+
|
| 75 |
+
- train_cross_entropy_epoch: 0.39623069763183594
|
| 76 |
+
|
| 77 |
+
- test_accuracy: 0.863082766532898
|
| 78 |
+
|
| 79 |
+
- test_matthewscorrcoef: 0.816676676273346
|
| 80 |
+
|
| 81 |
+
- test_f1score: 0.8613550662994385
|
| 82 |
+
|
| 83 |
+
- test_cross_entropy: 0.3769755959510803
|
| 84 |
+
|
| 85 |
+
- date_run: Apr-21-2022_t-02
|
| 86 |
+
|
| 87 |
+
- huggingface_tag: microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext
|