Training in progress, epoch 1
Browse files
README.md
ADDED
@@ -0,0 +1,104 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- conll2003
|
7 |
+
metrics:
|
8 |
+
- precision
|
9 |
+
- recall
|
10 |
+
- f1
|
11 |
+
- accuracy
|
12 |
+
model-index:
|
13 |
+
- name: bert-to-distilbert-NER
|
14 |
+
results:
|
15 |
+
- task:
|
16 |
+
name: Token Classification
|
17 |
+
type: token-classification
|
18 |
+
dataset:
|
19 |
+
name: conll2003
|
20 |
+
type: conll2003
|
21 |
+
config: conll2003
|
22 |
+
split: train
|
23 |
+
args: conll2003
|
24 |
+
metrics:
|
25 |
+
- name: Precision
|
26 |
+
type: precision
|
27 |
+
value: 0.014729299363057325
|
28 |
+
- name: Recall
|
29 |
+
type: recall
|
30 |
+
value: 0.018680578929653316
|
31 |
+
- name: F1
|
32 |
+
type: f1
|
33 |
+
value: 0.016471286541029827
|
34 |
+
- name: Accuracy
|
35 |
+
type: accuracy
|
36 |
+
value: 0.7599340672278802
|
37 |
+
---
|
38 |
+
|
39 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
40 |
+
should probably proofread and complete it, then remove this comment. -->
|
41 |
+
|
42 |
+
# bert-to-distilbert-NER
|
43 |
+
|
44 |
+
This model is a fine-tuned version of [dslim/bert-base-NER](https://huggingface.co/dslim/bert-base-NER) on the conll2003 dataset.
|
45 |
+
It achieves the following results on the evaluation set:
|
46 |
+
- Loss: 43.2398
|
47 |
+
- Precision: 0.0147
|
48 |
+
- Recall: 0.0187
|
49 |
+
- F1: 0.0165
|
50 |
+
- Accuracy: 0.7599
|
51 |
+
|
52 |
+
## Model description
|
53 |
+
|
54 |
+
More information needed
|
55 |
+
|
56 |
+
## Intended uses & limitations
|
57 |
+
|
58 |
+
More information needed
|
59 |
+
|
60 |
+
## Training and evaluation data
|
61 |
+
|
62 |
+
More information needed
|
63 |
+
|
64 |
+
## Training procedure
|
65 |
+
|
66 |
+
### Training hyperparameters
|
67 |
+
|
68 |
+
The following hyperparameters were used during training:
|
69 |
+
- learning_rate: 6e-05
|
70 |
+
- train_batch_size: 128
|
71 |
+
- eval_batch_size: 128
|
72 |
+
- seed: 33
|
73 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
74 |
+
- lr_scheduler_type: linear
|
75 |
+
- num_epochs: 15
|
76 |
+
- mixed_precision_training: Native AMP
|
77 |
+
|
78 |
+
### Training results
|
79 |
+
|
80 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
81 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
82 |
+
| 190.2685 | 1.0 | 110 | 127.2351 | 0.0157 | 0.0098 | 0.0120 | 0.7569 |
|
83 |
+
| 105.4389 | 2.0 | 220 | 97.1100 | 0.0281 | 0.0298 | 0.0289 | 0.7587 |
|
84 |
+
| 77.0337 | 3.0 | 330 | 76.9433 | 0.0136 | 0.0173 | 0.0152 | 0.7615 |
|
85 |
+
| 60.3477 | 4.0 | 440 | 65.9181 | 0.0130 | 0.0158 | 0.0143 | 0.7603 |
|
86 |
+
| 50.4086 | 5.0 | 550 | 58.5255 | 0.0170 | 0.0220 | 0.0192 | 0.7603 |
|
87 |
+
| 43.298 | 6.0 | 660 | 54.5405 | 0.0144 | 0.0187 | 0.0163 | 0.7594 |
|
88 |
+
| 39.0911 | 7.0 | 770 | 52.4767 | 0.0155 | 0.0195 | 0.0172 | 0.7613 |
|
89 |
+
| 35.07 | 8.0 | 880 | 49.1975 | 0.0170 | 0.0219 | 0.0192 | 0.7602 |
|
90 |
+
| 32.215 | 9.0 | 990 | 47.4422 | 0.0144 | 0.0187 | 0.0163 | 0.7599 |
|
91 |
+
| 29.9923 | 10.0 | 1100 | 46.5558 | 0.0167 | 0.0212 | 0.0187 | 0.7606 |
|
92 |
+
| 28.3599 | 11.0 | 1210 | 45.6301 | 0.0171 | 0.0214 | 0.0190 | 0.7613 |
|
93 |
+
| 26.8163 | 12.0 | 1320 | 45.0483 | 0.0141 | 0.0177 | 0.0157 | 0.7606 |
|
94 |
+
| 25.7434 | 13.0 | 1430 | 44.0639 | 0.0176 | 0.0222 | 0.0196 | 0.7605 |
|
95 |
+
| 24.9853 | 14.0 | 1540 | 43.6618 | 0.0148 | 0.0187 | 0.0165 | 0.7606 |
|
96 |
+
| 24.3179 | 15.0 | 1650 | 43.2398 | 0.0147 | 0.0187 | 0.0165 | 0.7599 |
|
97 |
+
|
98 |
+
|
99 |
+
### Framework versions
|
100 |
+
|
101 |
+
- Transformers 4.25.1
|
102 |
+
- Pytorch 1.13.1+cu116
|
103 |
+
- Datasets 2.8.0
|
104 |
+
- Tokenizers 0.13.2
|
logs/1674367746.1589544/events.out.tfevents.1674367746.6e8f83ae5915.190.3
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a471ab48be7dca22771a2c81885d001d571956c91e334440b7b1f67d6ede1424
|
3 |
+
size 5522
|
logs/events.out.tfevents.1674366449.6e8f83ae5915.190.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:84fe6225bea0e548ab56ddbc7450686f69a433e7562b41a783fbf691d1f26425
|
3 |
+
size 13869
|
logs/events.out.tfevents.1674367746.6e8f83ae5915.190.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dda2a997347c766402349f66f74ca268f035ae816fc3a183ae1739dd92279b1c
|
3 |
+
size 4724
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 260836653
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e0191325d961e6c64e48558b8c45ceb64b74b16ebe26d305eb45fcd43f1099c1
|
3 |
size 260836653
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3515
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3eb1c12b7e569f5d26d17f3df0041ef3764f38456b366b0d366e544f6648c068
|
3 |
size 3515
|