update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- precision
|
7 |
+
- recall
|
8 |
+
- f1
|
9 |
+
- accuracy
|
10 |
+
model-index:
|
11 |
+
- name: Medical-NER-finetuned-ner
|
12 |
+
results: []
|
13 |
+
---
|
14 |
+
|
15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
16 |
+
should probably proofread and complete it, then remove this comment. -->
|
17 |
+
|
18 |
+
# Medical-NER-finetuned-ner
|
19 |
+
|
20 |
+
This model is a fine-tuned version of [Clinical-AI-Apollo/Medical-NER](https://huggingface.co/Clinical-AI-Apollo/Medical-NER) on an unknown dataset.
|
21 |
+
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.3114
|
23 |
+
- Precision: 0.7903
|
24 |
+
- Recall: 0.9005
|
25 |
+
- F1: 0.8418
|
26 |
+
- Accuracy: 0.9313
|
27 |
+
|
28 |
+
## Model description
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Intended uses & limitations
|
33 |
+
|
34 |
+
More information needed
|
35 |
+
|
36 |
+
## Training and evaluation data
|
37 |
+
|
38 |
+
More information needed
|
39 |
+
|
40 |
+
## Training procedure
|
41 |
+
|
42 |
+
### Training hyperparameters
|
43 |
+
|
44 |
+
The following hyperparameters were used during training:
|
45 |
+
- learning_rate: 2e-05
|
46 |
+
- train_batch_size: 4
|
47 |
+
- eval_batch_size: 4
|
48 |
+
- seed: 42
|
49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
50 |
+
- lr_scheduler_type: cosine
|
51 |
+
- num_epochs: 20
|
52 |
+
|
53 |
+
### Training results
|
54 |
+
|
55 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
56 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
57 |
+
| No log | 1.0 | 90 | 0.9174 | 0.4239 | 0.3613 | 0.3901 | 0.7448 |
|
58 |
+
| No log | 2.0 | 180 | 0.6814 | 0.5257 | 0.5521 | 0.5386 | 0.7899 |
|
59 |
+
| No log | 3.0 | 270 | 0.6262 | 0.5383 | 0.7265 | 0.6184 | 0.7974 |
|
60 |
+
| No log | 4.0 | 360 | 0.4934 | 0.6065 | 0.7291 | 0.6622 | 0.8434 |
|
61 |
+
| No log | 5.0 | 450 | 0.5071 | 0.6102 | 0.7946 | 0.6903 | 0.8431 |
|
62 |
+
| 0.7847 | 6.0 | 540 | 0.4195 | 0.6863 | 0.7963 | 0.7372 | 0.8744 |
|
63 |
+
| 0.7847 | 7.0 | 630 | 0.4215 | 0.6850 | 0.8386 | 0.7541 | 0.8816 |
|
64 |
+
| 0.7847 | 8.0 | 720 | 0.3807 | 0.7287 | 0.8440 | 0.7822 | 0.8985 |
|
65 |
+
| 0.7847 | 9.0 | 810 | 0.3474 | 0.7383 | 0.8479 | 0.7893 | 0.9079 |
|
66 |
+
| 0.7847 | 10.0 | 900 | 0.3259 | 0.7583 | 0.8679 | 0.8094 | 0.9135 |
|
67 |
+
| 0.7847 | 11.0 | 990 | 0.3428 | 0.7595 | 0.8812 | 0.8158 | 0.9151 |
|
68 |
+
| 0.2288 | 12.0 | 1080 | 0.3469 | 0.7568 | 0.8821 | 0.8147 | 0.9147 |
|
69 |
+
| 0.2288 | 13.0 | 1170 | 0.3211 | 0.7790 | 0.8880 | 0.8299 | 0.9257 |
|
70 |
+
| 0.2288 | 14.0 | 1260 | 0.3217 | 0.7847 | 0.8909 | 0.8344 | 0.9271 |
|
71 |
+
| 0.2288 | 15.0 | 1350 | 0.2944 | 0.7952 | 0.8941 | 0.8418 | 0.9321 |
|
72 |
+
| 0.2288 | 16.0 | 1440 | 0.3244 | 0.7822 | 0.8986 | 0.8364 | 0.9275 |
|
73 |
+
| 0.1273 | 17.0 | 1530 | 0.3153 | 0.7911 | 0.9012 | 0.8426 | 0.9307 |
|
74 |
+
| 0.1273 | 18.0 | 1620 | 0.3198 | 0.7874 | 0.9005 | 0.8402 | 0.9298 |
|
75 |
+
| 0.1273 | 19.0 | 1710 | 0.3109 | 0.7911 | 0.9012 | 0.8426 | 0.9315 |
|
76 |
+
| 0.1273 | 20.0 | 1800 | 0.3114 | 0.7903 | 0.9005 | 0.8418 | 0.9313 |
|
77 |
+
|
78 |
+
|
79 |
+
### Framework versions
|
80 |
+
|
81 |
+
- Transformers 4.30.0
|
82 |
+
- Pytorch 2.2.1+cu121
|
83 |
+
- Datasets 2.18.0
|
84 |
+
- Tokenizers 0.13.3
|