update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- precision
|
7 |
+
- recall
|
8 |
+
- f1
|
9 |
+
- accuracy
|
10 |
+
model-index:
|
11 |
+
- name: balanced-augmented-bert-large-gest-pred-seqeval-partialmatch-2
|
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 |
+
# balanced-augmented-bert-large-gest-pred-seqeval-partialmatch-2
|
19 |
+
|
20 |
+
This model is a fine-tuned version of [bert-large-cased](https://huggingface.co/bert-large-cased) on the None dataset.
|
21 |
+
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.5066
|
23 |
+
- Precision: 0.9246
|
24 |
+
- Recall: 0.9232
|
25 |
+
- F1: 0.9156
|
26 |
+
- Accuracy: 0.9045
|
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: 16
|
47 |
+
- eval_batch_size: 16
|
48 |
+
- seed: 42
|
49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
50 |
+
- lr_scheduler_type: linear
|
51 |
+
- num_epochs: 20
|
52 |
+
|
53 |
+
### Training results
|
54 |
+
|
55 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
56 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
57 |
+
| 2.9285 | 1.0 | 52 | 2.4277 | 0.2289 | 0.1482 | 0.1320 | 0.3077 |
|
58 |
+
| 1.9821 | 2.0 | 104 | 1.6710 | 0.5805 | 0.4849 | 0.4419 | 0.5576 |
|
59 |
+
| 1.3372 | 3.0 | 156 | 1.1456 | 0.6591 | 0.6279 | 0.6073 | 0.6703 |
|
60 |
+
| 0.8553 | 4.0 | 208 | 0.7989 | 0.7950 | 0.7956 | 0.7738 | 0.7864 |
|
61 |
+
| 0.4964 | 5.0 | 260 | 0.6153 | 0.8422 | 0.8447 | 0.8281 | 0.8378 |
|
62 |
+
| 0.2985 | 6.0 | 312 | 0.4399 | 0.9124 | 0.8982 | 0.8966 | 0.8814 |
|
63 |
+
| 0.1825 | 7.0 | 364 | 0.4938 | 0.8936 | 0.9034 | 0.8883 | 0.8829 |
|
64 |
+
| 0.1178 | 8.0 | 416 | 0.4713 | 0.9087 | 0.9188 | 0.9069 | 0.8912 |
|
65 |
+
| 0.0812 | 9.0 | 468 | 0.4012 | 0.9108 | 0.9274 | 0.9130 | 0.9045 |
|
66 |
+
| 0.0579 | 10.0 | 520 | 0.4695 | 0.9120 | 0.9132 | 0.9050 | 0.8942 |
|
67 |
+
| 0.0345 | 11.0 | 572 | 0.5327 | 0.9196 | 0.9165 | 0.9083 | 0.8976 |
|
68 |
+
| 0.0309 | 12.0 | 624 | 0.5243 | 0.9273 | 0.9207 | 0.9146 | 0.9025 |
|
69 |
+
| 0.0234 | 13.0 | 676 | 0.5089 | 0.9271 | 0.9243 | 0.9165 | 0.8996 |
|
70 |
+
| 0.0175 | 14.0 | 728 | 0.4750 | 0.9284 | 0.9258 | 0.9190 | 0.9059 |
|
71 |
+
| 0.015 | 15.0 | 780 | 0.4891 | 0.9310 | 0.9277 | 0.9210 | 0.9079 |
|
72 |
+
| 0.0109 | 16.0 | 832 | 0.5126 | 0.9240 | 0.9222 | 0.9153 | 0.9045 |
|
73 |
+
| 0.0085 | 17.0 | 884 | 0.4512 | 0.9320 | 0.9315 | 0.9246 | 0.9123 |
|
74 |
+
| 0.0077 | 18.0 | 936 | 0.5363 | 0.9241 | 0.9226 | 0.9149 | 0.9035 |
|
75 |
+
| 0.0058 | 19.0 | 988 | 0.5033 | 0.9246 | 0.9232 | 0.9156 | 0.9045 |
|
76 |
+
| 0.0062 | 20.0 | 1040 | 0.5066 | 0.9246 | 0.9232 | 0.9156 | 0.9045 |
|
77 |
+
|
78 |
+
|
79 |
+
### Framework versions
|
80 |
+
|
81 |
+
- Transformers 4.27.3
|
82 |
+
- Pytorch 1.13.1+cu116
|
83 |
+
- Datasets 2.10.1
|
84 |
+
- Tokenizers 0.13.2
|