End of training
Browse files
README.md
ADDED
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- generated_from_trainer
|
4 |
+
datasets:
|
5 |
+
- glue
|
6 |
+
metrics:
|
7 |
+
- accuracy
|
8 |
+
model-index:
|
9 |
+
- name: tiny-bert-mnli-distilled
|
10 |
+
results:
|
11 |
+
- task:
|
12 |
+
name: Text Classification
|
13 |
+
type: text-classification
|
14 |
+
dataset:
|
15 |
+
name: glue
|
16 |
+
type: glue
|
17 |
+
args: mnli
|
18 |
+
metrics:
|
19 |
+
- name: Accuracy
|
20 |
+
type: accuracy
|
21 |
+
value: 0.5818644931227712
|
22 |
+
---
|
23 |
+
|
24 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
25 |
+
should probably proofread and complete it, then remove this comment. -->
|
26 |
+
|
27 |
+
# tiny-bert-mnli-distilled
|
28 |
+
|
29 |
+
This model is a fine-tuned version of [M-FAC/bert-mini-finetuned-mnli](https://huggingface.co/M-FAC/bert-mini-finetuned-mnli) on the glue dataset.
|
30 |
+
It achieves the following results on the evaluation set:
|
31 |
+
- Loss: 1.5018
|
32 |
+
- Accuracy: 0.5819
|
33 |
+
- F1 score: 0.5782
|
34 |
+
- Precision score: 0.6036
|
35 |
+
- Metric recall: 0.5819
|
36 |
+
|
37 |
+
## Model description
|
38 |
+
|
39 |
+
More information needed
|
40 |
+
|
41 |
+
## Intended uses & limitations
|
42 |
+
|
43 |
+
More information needed
|
44 |
+
|
45 |
+
## Training and evaluation data
|
46 |
+
|
47 |
+
More information needed
|
48 |
+
|
49 |
+
## Training procedure
|
50 |
+
|
51 |
+
### Training hyperparameters
|
52 |
+
|
53 |
+
The following hyperparameters were used during training:
|
54 |
+
- learning_rate: 0.0005
|
55 |
+
- train_batch_size: 64
|
56 |
+
- eval_batch_size: 32
|
57 |
+
- seed: 33
|
58 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
59 |
+
- lr_scheduler_type: linear
|
60 |
+
- num_epochs: 4
|
61 |
+
- mixed_precision_training: Native AMP
|
62 |
+
|
63 |
+
### Training results
|
64 |
+
|
65 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 score | Precision score | Metric recall |
|
66 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:---------------:|:-------------:|
|
67 |
+
| 1.4475 | 1.0 | 614 | 1.4296 | 0.4521 | 0.4070 | 0.5621 | 0.4521 |
|
68 |
+
| 1.3354 | 2.0 | 1228 | 1.4320 | 0.4805 | 0.4579 | 0.5276 | 0.4805 |
|
69 |
+
| 1.2244 | 3.0 | 1842 | 1.4786 | 0.5699 | 0.5602 | 0.5865 | 0.5699 |
|
70 |
+
| 1.1416 | 4.0 | 2456 | 1.5018 | 0.5819 | 0.5782 | 0.6036 | 0.5819 |
|
71 |
+
|
72 |
+
|
73 |
+
### Framework versions
|
74 |
+
|
75 |
+
- Transformers 4.18.0
|
76 |
+
- Pytorch 1.9.1
|
77 |
+
- Datasets 2.1.0
|
78 |
+
- Tokenizers 0.11.6
|
logs/events.out.tfevents.1650255678.cda890f99651.34.2
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 sha256:5564319120f0841fb378bf67cb550201e3388ac0b9817d2adcc9547d77fdc292
|
3 |
+
size 6604
|