update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- generated_from_trainer
|
4 |
+
datasets:
|
5 |
+
- glue
|
6 |
+
metrics:
|
7 |
+
- accuracy
|
8 |
+
model-index:
|
9 |
+
- name: hBERTv2_new_pretrain_48_KD_sst2
|
10 |
+
results:
|
11 |
+
- task:
|
12 |
+
name: Text Classification
|
13 |
+
type: text-classification
|
14 |
+
dataset:
|
15 |
+
name: glue
|
16 |
+
type: glue
|
17 |
+
config: sst2
|
18 |
+
split: validation
|
19 |
+
args: sst2
|
20 |
+
metrics:
|
21 |
+
- name: Accuracy
|
22 |
+
type: accuracy
|
23 |
+
value: 0.8107798165137615
|
24 |
+
---
|
25 |
+
|
26 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
27 |
+
should probably proofread and complete it, then remove this comment. -->
|
28 |
+
|
29 |
+
# hBERTv2_new_pretrain_48_KD_sst2
|
30 |
+
|
31 |
+
This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48_KD](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48_KD) on the glue dataset.
|
32 |
+
It achieves the following results on the evaluation set:
|
33 |
+
- Loss: 0.6627
|
34 |
+
- Accuracy: 0.8108
|
35 |
+
|
36 |
+
## Model description
|
37 |
+
|
38 |
+
More information needed
|
39 |
+
|
40 |
+
## Intended uses & limitations
|
41 |
+
|
42 |
+
More information needed
|
43 |
+
|
44 |
+
## Training and evaluation data
|
45 |
+
|
46 |
+
More information needed
|
47 |
+
|
48 |
+
## Training procedure
|
49 |
+
|
50 |
+
### Training hyperparameters
|
51 |
+
|
52 |
+
The following hyperparameters were used during training:
|
53 |
+
- learning_rate: 4e-05
|
54 |
+
- train_batch_size: 128
|
55 |
+
- eval_batch_size: 128
|
56 |
+
- seed: 10
|
57 |
+
- distributed_type: multi-GPU
|
58 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
59 |
+
- lr_scheduler_type: linear
|
60 |
+
- num_epochs: 50
|
61 |
+
|
62 |
+
### Training results
|
63 |
+
|
64 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
65 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
66 |
+
| 0.6251 | 1.0 | 527 | 0.4825 | 0.7787 |
|
67 |
+
| 0.2866 | 2.0 | 1054 | 0.6289 | 0.8073 |
|
68 |
+
| 0.2223 | 3.0 | 1581 | 0.4860 | 0.8050 |
|
69 |
+
| 0.1929 | 4.0 | 2108 | 0.5174 | 0.8108 |
|
70 |
+
| 0.1698 | 5.0 | 2635 | 0.4868 | 0.8050 |
|
71 |
+
| 0.1531 | 6.0 | 3162 | 0.6627 | 0.8108 |
|
72 |
+
|
73 |
+
|
74 |
+
### Framework versions
|
75 |
+
|
76 |
+
- Transformers 4.30.2
|
77 |
+
- Pytorch 1.14.0a0+410ce96
|
78 |
+
- Datasets 2.12.0
|
79 |
+
- Tokenizers 0.13.3
|