File size: 3,470 Bytes
73e5a3b
 
 
27eeb55
 
73e5a3b
 
 
 
27eeb55
 
 
 
 
 
 
 
 
 
 
73e5a3b
 
 
 
 
 
 
27eeb55
73e5a3b
27eeb55
 
73e5a3b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
---
tags:
- generated_from_trainer
datasets:
- gokuls/wiki_book_corpus_complete_processed_bert_dataset
metrics:
- accuracy
model-index:
- name: HBERTv1_emb_compress_48_L12_H512_A8
  results:
  - task:
      name: Masked Language Modeling
      type: fill-mask
    dataset:
      name: gokuls/wiki_book_corpus_complete_processed_bert_dataset
      type: gokuls/wiki_book_corpus_complete_processed_bert_dataset
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.15118347474846977
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# HBERTv1_emb_compress_48_L12_H512_A8

This model is a fine-tuned version of [](https://huggingface.co/) on the gokuls/wiki_book_corpus_complete_processed_bert_dataset dataset.
It achieves the following results on the evaluation set:
- Loss: 6.0426
- Accuracy: 0.1512

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 56
- eval_batch_size: 56
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10000
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step   | Validation Loss | Accuracy |
|:-------------:|:-----:|:------:|:---------------:|:--------:|
| 7.102         | 0.1   | 10000  | 7.0825          | 0.0834   |
| 6.6803        | 0.19  | 20000  | 6.6756          | 0.1066   |
| 6.5348        | 0.29  | 30000  | 6.5298          | 0.1196   |
| 6.4394        | 0.38  | 40000  | 6.4389          | 0.1274   |
| 6.3686        | 0.48  | 50000  | 6.3726          | 0.1332   |
| 6.3213        | 0.57  | 60000  | 6.3189          | 0.1358   |
| 6.281         | 0.67  | 70000  | 6.2812          | 0.1382   |
| 6.2506        | 0.76  | 80000  | 6.2467          | 0.1401   |
| 6.221         | 0.86  | 90000  | 6.2216          | 0.1423   |
| 6.206         | 0.96  | 100000 | 6.1978          | 0.1431   |
| 6.1831        | 1.05  | 110000 | 6.1796          | 0.1449   |
| 6.1609        | 1.15  | 120000 | 6.1630          | 0.1457   |
| 6.153         | 1.24  | 130000 | 6.1505          | 0.1464   |
| 6.142         | 1.34  | 140000 | 6.1380          | 0.1471   |
| 6.1281        | 1.43  | 150000 | 6.1257          | 0.1477   |
| 6.1173        | 1.53  | 160000 | 6.1173          | 0.1481   |
| 6.1102        | 1.62  | 170000 | 6.1083          | 0.1489   |
| 6.1011        | 1.72  | 180000 | 6.1001          | 0.1487   |
| 6.0869        | 1.82  | 190000 | 6.0933          | 0.1493   |
| 6.0838        | 1.91  | 200000 | 6.0864          | 0.1494   |
| 6.0745        | 2.01  | 210000 | 6.0805          | 0.1499   |
| 6.0757        | 2.1   | 220000 | 6.0723          | 0.1503   |
| 6.0695        | 2.2   | 230000 | 6.0701          | 0.1502   |
| 6.0595        | 2.29  | 240000 | 6.0623          | 0.1506   |
| 6.0579        | 2.39  | 250000 | 6.0582          | 0.1506   |
| 6.0534        | 2.49  | 260000 | 6.0526          | 0.1509   |
| 6.0465        | 2.58  | 270000 | 6.0433          | 0.1510   |


### Framework versions

- Transformers 4.33.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.14.5
- Tokenizers 0.13.3