Christopher McMaster
commited on
Commit
•
f004d37
1
Parent(s):
3a7d0a9
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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: adr-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 |
+
# adr-ner
|
19 |
+
|
20 |
+
This model is a fine-tuned version of [austin/Austin-MeDeBERTa](https://huggingface.co/austin/Austin-MeDeBERTa) on the None dataset.
|
21 |
+
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.0434
|
23 |
+
- Precision: 0.7305
|
24 |
+
- Recall: 0.6934
|
25 |
+
- F1: 0.7115
|
26 |
+
- Accuracy: 0.9941
|
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: 5e-05
|
46 |
+
- train_batch_size: 12
|
47 |
+
- eval_batch_size: 12
|
48 |
+
- seed: 42
|
49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
50 |
+
- lr_scheduler_type: linear
|
51 |
+
- num_epochs: 15
|
52 |
+
|
53 |
+
### Training results
|
54 |
+
|
55 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
56 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
57 |
+
| No log | 1.0 | 107 | 0.0630 | 0.0 | 0.0 | 0.0 | 0.9876 |
|
58 |
+
| No log | 2.0 | 214 | 0.0308 | 0.4282 | 0.3467 | 0.3832 | 0.9900 |
|
59 |
+
| No log | 3.0 | 321 | 0.0254 | 0.5544 | 0.5603 | 0.5573 | 0.9920 |
|
60 |
+
| No log | 4.0 | 428 | 0.0280 | 0.6430 | 0.5751 | 0.6071 | 0.9929 |
|
61 |
+
| 0.0465 | 5.0 | 535 | 0.0266 | 0.5348 | 0.7146 | 0.6118 | 0.9915 |
|
62 |
+
| 0.0465 | 6.0 | 642 | 0.0423 | 0.7632 | 0.5793 | 0.6587 | 0.9939 |
|
63 |
+
| 0.0465 | 7.0 | 749 | 0.0336 | 0.6957 | 0.6765 | 0.6860 | 0.9939 |
|
64 |
+
| 0.0465 | 8.0 | 856 | 0.0370 | 0.6876 | 0.6702 | 0.6788 | 0.9936 |
|
65 |
+
| 0.0465 | 9.0 | 963 | 0.0349 | 0.6555 | 0.7040 | 0.6789 | 0.9932 |
|
66 |
+
| 0.0044 | 10.0 | 1070 | 0.0403 | 0.6910 | 0.6808 | 0.6858 | 0.9938 |
|
67 |
+
| 0.0044 | 11.0 | 1177 | 0.0415 | 0.7140 | 0.6808 | 0.6970 | 0.9939 |
|
68 |
+
| 0.0044 | 12.0 | 1284 | 0.0440 | 0.7349 | 0.6681 | 0.6999 | 0.9941 |
|
69 |
+
| 0.0044 | 13.0 | 1391 | 0.0423 | 0.7097 | 0.6977 | 0.7036 | 0.9941 |
|
70 |
+
| 0.0044 | 14.0 | 1498 | 0.0435 | 0.7174 | 0.6977 | 0.7074 | 0.9941 |
|
71 |
+
| 0.0006 | 15.0 | 1605 | 0.0434 | 0.7305 | 0.6934 | 0.7115 | 0.9941 |
|
72 |
+
|
73 |
+
|
74 |
+
### Framework versions
|
75 |
+
|
76 |
+
- Transformers 4.14.1
|
77 |
+
- Pytorch 1.10.0+cu113
|
78 |
+
- Datasets 1.16.1
|
79 |
+
- Tokenizers 0.10.3
|