Commit
•
c3a27a2
1
Parent(s):
059fbe2
End of training
Browse files
README.md
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: medicalai/ClinicalBERT
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: CRAFT_ClinicalBERT_NER
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# CRAFT_ClinicalBERT_NER
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [medicalai/ClinicalBERT](https://huggingface.co/medicalai/ClinicalBERT) on the None dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 0.1733
|
| 18 |
+
- Seqeval classification report: precision recall f1-score support
|
| 19 |
+
|
| 20 |
+
CHEBI 0.68 0.66 0.67 1365
|
| 21 |
+
CL 0.55 0.50 0.52 284
|
| 22 |
+
GGP 0.87 0.81 0.84 4632
|
| 23 |
+
GO 0.66 0.65 0.65 8852
|
| 24 |
+
SO 0.68 0.50 0.58 616
|
| 25 |
+
Taxon 0.81 0.73 0.77 986
|
| 26 |
+
|
| 27 |
+
micro avg 0.72 0.69 0.71 16735
|
| 28 |
+
macro avg 0.71 0.64 0.67 16735
|
| 29 |
+
weighted avg 0.73 0.69 0.71 16735
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
## Model description
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Intended uses & limitations
|
| 37 |
+
|
| 38 |
+
More information needed
|
| 39 |
+
|
| 40 |
+
## Training and evaluation data
|
| 41 |
+
|
| 42 |
+
More information needed
|
| 43 |
+
|
| 44 |
+
## Training procedure
|
| 45 |
+
|
| 46 |
+
### Training hyperparameters
|
| 47 |
+
|
| 48 |
+
The following hyperparameters were used during training:
|
| 49 |
+
- learning_rate: 2e-05
|
| 50 |
+
- train_batch_size: 16
|
| 51 |
+
- eval_batch_size: 16
|
| 52 |
+
- seed: 42
|
| 53 |
+
- gradient_accumulation_steps: 2
|
| 54 |
+
- total_train_batch_size: 32
|
| 55 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 56 |
+
- lr_scheduler_type: linear
|
| 57 |
+
- num_epochs: 3
|
| 58 |
+
|
| 59 |
+
### Training results
|
| 60 |
+
|
| 61 |
+
| Training Loss | Epoch | Step | Validation Loss | Seqeval classification report |
|
| 62 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|
|
| 63 |
+
| No log | 1.0 | 347 | 0.1894 | precision recall f1-score support
|
| 64 |
+
|
| 65 |
+
CHEBI 0.64 0.56 0.60 1365
|
| 66 |
+
CL 0.53 0.35 0.42 284
|
| 67 |
+
GGP 0.84 0.77 0.81 4632
|
| 68 |
+
GO 0.60 0.61 0.60 8852
|
| 69 |
+
SO 0.53 0.46 0.49 616
|
| 70 |
+
Taxon 0.78 0.66 0.71 986
|
| 71 |
+
|
| 72 |
+
micro avg 0.68 0.64 0.66 16735
|
| 73 |
+
macro avg 0.65 0.57 0.61 16735
|
| 74 |
+
weighted avg 0.68 0.64 0.66 16735
|
| 75 |
+
|
|
| 76 |
+
| 0.2231 | 2.0 | 695 | 0.1740 | precision recall f1-score support
|
| 77 |
+
|
| 78 |
+
CHEBI 0.69 0.63 0.66 1365
|
| 79 |
+
CL 0.56 0.44 0.49 284
|
| 80 |
+
GGP 0.83 0.79 0.81 4632
|
| 81 |
+
GO 0.65 0.65 0.65 8852
|
| 82 |
+
SO 0.68 0.47 0.55 616
|
| 83 |
+
Taxon 0.81 0.72 0.76 986
|
| 84 |
+
|
| 85 |
+
micro avg 0.71 0.68 0.69 16735
|
| 86 |
+
macro avg 0.70 0.62 0.65 16735
|
| 87 |
+
weighted avg 0.71 0.68 0.69 16735
|
| 88 |
+
|
|
| 89 |
+
| 0.0813 | 3.0 | 1041 | 0.1733 | precision recall f1-score support
|
| 90 |
+
|
| 91 |
+
CHEBI 0.68 0.66 0.67 1365
|
| 92 |
+
CL 0.55 0.50 0.52 284
|
| 93 |
+
GGP 0.87 0.81 0.84 4632
|
| 94 |
+
GO 0.66 0.65 0.65 8852
|
| 95 |
+
SO 0.68 0.50 0.58 616
|
| 96 |
+
Taxon 0.81 0.73 0.77 986
|
| 97 |
+
|
| 98 |
+
micro avg 0.72 0.69 0.71 16735
|
| 99 |
+
macro avg 0.71 0.64 0.67 16735
|
| 100 |
+
weighted avg 0.73 0.69 0.71 16735
|
| 101 |
+
|
|
| 102 |
+
|
| 103 |
+
|
| 104 |
+
### Framework versions
|
| 105 |
+
|
| 106 |
+
- Transformers 4.35.2
|
| 107 |
+
- Pytorch 2.1.0+cu121
|
| 108 |
+
- Datasets 2.15.0
|
| 109 |
+
- Tokenizers 0.15.0
|