Hugging Face's logo

jamie613
/
custom_BERT_NER

Token Classification
Transformers
TensorBoard
Safetensors
Chinese
bert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
jamie613 commited on
Commit
ee1e337
•
1 Parent(s): e62357e

dataset=155, epochs=50, batch_size=1, early_stopping=eval_precision

Browse files
Files changed (3) hide show
  1. README.md +24 -19
  2. model.safetensors +1 -1
  3. tokenizer.json +1 -10
README.md CHANGED
@@ -20,17 +20,17 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) on an unknown dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.3270
24
- - Perf P: 0.9
25
- - Perf R: 0.9529
26
- - Inst P: 0.8657
27
- - Inst R: 0.8657
28
- - Comp P: 0.9341
29
- - Comp R: 0.9341
30
- - Precision: 0.8256
31
- - Recall: 0.8311
32
- - F1: 0.8283
33
- - Accuracy: 0.9288
34
 
35
  ## Model description
36
 
@@ -61,14 +61,19 @@ The following hyperparameters were used during training:
61
 
62
  | Training Loss | Epoch | Step | Validation Loss | Perf P | Perf R | Inst P | Inst R | Comp P | Comp R | Precision | Recall | F1 | Accuracy |
63
  |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:------:|:------:|:------:|:---------:|:------:|:------:|:--------:|
64
- | 0.9205 | 1.0 | 135 | 0.4005 | 0.8148 | 0.7765 | 0.6923 | 0.8060 | 0.8101 | 0.7033 | 0.7042 | 0.6488 | 0.6754 | 0.8767 |
65
- | 0.2812 | 2.0 | 270 | 0.2675 | 0.8462 | 0.9059 | 0.8485 | 0.8358 | 0.8646 | 0.9121 | 0.7841 | 0.8077 | 0.7957 | 0.9235 |
66
- | 0.1573 | 3.0 | 405 | 0.2843 | 0.8778 | 0.9294 | 0.9048 | 0.8507 | 0.9014 | 0.7033 | 0.7713 | 0.7559 | 0.7635 | 0.9106 |
67
- | 0.1013 | 4.0 | 540 | 0.2547 | 0.8316 | 0.9294 | 0.8026 | 0.9104 | 0.8630 | 0.6923 | 0.7465 | 0.7926 | 0.7689 | 0.9235 |
68
- | 0.0688 | 5.0 | 675 | 0.2390 | 0.8333 | 0.9412 | 0.8611 | 0.9254 | 0.8690 | 0.8022 | 0.7977 | 0.8043 | 0.8010 | 0.9321 |
69
- | 0.0499 | 6.0 | 810 | 0.2709 | 0.8571 | 0.9176 | 0.8939 | 0.8806 | 0.8438 | 0.8901 | 0.7932 | 0.8211 | 0.8069 | 0.9327 |
70
- | 0.0387 | 7.0 | 945 | 0.3308 | 0.8941 | 0.8941 | 0.7532 | 0.8657 | 0.9178 | 0.7363 | 0.7638 | 0.7625 | 0.7632 | 0.9168 |
71
- | 0.0254 | 8.0 | 1080 | 0.3270 | 0.9 | 0.9529 | 0.8657 | 0.8657 | 0.9341 | 0.9341 | 0.8256 | 0.8311 | 0.8283 | 0.9288 |
 
 
 
 
 
72
 
73
 
74
  ### Framework versions
 
20
 
21
  This model is a fine-tuned version of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) on an unknown dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.2331
24
+ - Perf P: 0.92
25
+ - Perf R: 0.9583
26
+ - Inst P: 0.9322
27
+ - Inst R: 0.8594
28
+ - Comp P: 0.9060
29
+ - Comp R: 0.8908
30
+ - Precision: 0.8550
31
+ - Recall: 0.8550
32
+ - F1: 0.8550
33
+ - Accuracy: 0.9558
34
 
35
  ## Model description
36
 
 
61
 
62
  | Training Loss | Epoch | Step | Validation Loss | Perf P | Perf R | Inst P | Inst R | Comp P | Comp R | Precision | Recall | F1 | Accuracy |
63
  |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:------:|:------:|:------:|:---------:|:------:|:------:|:--------:|
64
+ | 0.8805 | 1.0 | 135 | 0.3680 | 0.7467 | 0.7778 | 0.5775 | 0.6406 | 0.8932 | 0.7731 | 0.7111 | 0.6154 | 0.6598 | 0.8865 |
65
+ | 0.2916 | 2.0 | 270 | 0.2304 | 0.8 | 0.8889 | 0.9231 | 0.75 | 0.9126 | 0.7899 | 0.7671 | 0.7648 | 0.7659 | 0.9306 |
66
+ | 0.1632 | 3.0 | 405 | 0.1824 | 0.8354 | 0.9167 | 0.8909 | 0.7656 | 0.9138 | 0.8908 | 0.8088 | 0.8195 | 0.8141 | 0.9441 |
67
+ | 0.0956 | 4.0 | 540 | 0.2044 | 0.7975 | 0.875 | 0.8308 | 0.8438 | 0.9615 | 0.8403 | 0.8104 | 0.8033 | 0.8068 | 0.9398 |
68
+ | 0.0634 | 5.0 | 675 | 0.2080 | 0.8519 | 0.9583 | 0.9464 | 0.8281 | 0.9135 | 0.7983 | 0.8387 | 0.8151 | 0.8267 | 0.9438 |
69
+ | 0.0446 | 6.0 | 810 | 0.2260 | 0.8625 | 0.9583 | 0.9355 | 0.9062 | 0.9316 | 0.9160 | 0.8559 | 0.8521 | 0.8540 | 0.9494 |
70
+ | 0.0307 | 7.0 | 945 | 0.1812 | 0.8608 | 0.9444 | 0.8429 | 0.9219 | 0.9304 | 0.8992 | 0.8582 | 0.8595 | 0.8588 | 0.9580 |
71
+ | 0.0224 | 8.0 | 1080 | 0.2485 | 0.7609 | 0.9722 | 0.9153 | 0.8438 | 0.9231 | 0.8067 | 0.8373 | 0.8225 | 0.8299 | 0.9411 |
72
+ | 0.0179 | 9.0 | 1215 | 0.2065 | 0.8625 | 0.9583 | 0.9153 | 0.8438 | 0.9450 | 0.8655 | 0.8582 | 0.8506 | 0.8544 | 0.9546 |
73
+ | 0.0189 | 10.0 | 1350 | 0.2236 | 0.8701 | 0.9306 | 0.9661 | 0.8906 | 0.9231 | 0.8067 | 0.8754 | 0.8624 | 0.8689 | 0.9574 |
74
+ | 0.0121 | 11.0 | 1485 | 0.2173 | 0.8313 | 0.9583 | 0.95 | 0.8906 | 0.9174 | 0.8403 | 0.8569 | 0.8595 | 0.8582 | 0.9534 |
75
+ | 0.0083 | 12.0 | 1620 | 0.2348 | 0.8023 | 0.9583 | 0.9180 | 0.875 | 0.9065 | 0.8151 | 0.8550 | 0.8462 | 0.8506 | 0.9543 |
76
+ | 0.0081 | 13.0 | 1755 | 0.2331 | 0.92 | 0.9583 | 0.9322 | 0.8594 | 0.9060 | 0.8908 | 0.8550 | 0.8550 | 0.8550 | 0.9558 |
77
 
78
 
79
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:32db6d6c8b725d9f66d76767a5086664b9cfd3aa43ca81baf6d85bcc76b12983
3
  size 709139348
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79e2ff74d8f62d76b324a47b3ad97a514bb16be8804bef5332f5e1d134d8b4b6
3
  size 709139348
tokenizer.json CHANGED
@@ -1,16 +1,7 @@
1
  {
2
  "version": "1.0",
3
  "truncation": null,
4
- "padding": {
5
- "strategy": {
6
- "Fixed": 512
7
- },
8
- "direction": "Right",
9
- "pad_to_multiple_of": null,
10
- "pad_id": 0,
11
- "pad_type_id": 0,
12
- "pad_token": "[PAD]"
13
- },
14
  "added_tokens": [
15
  {
16
  "id": 0,
 
1
  {
2
  "version": "1.0",
3
  "truncation": null,
4
+ "padding": null,
 
 
 
 
 
 
 
 
 
5
  "added_tokens": [
6
  {
7
  "id": 0,