File size: 3,660 Bytes
01e26db
 
 
b47335b
 
01e26db
 
 
 
b47335b
 
 
 
 
 
 
 
 
 
 
01e26db
 
 
 
 
 
 
b47335b
01e26db
b47335b
01e26db
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
100
101
102
---
tags:
- generated_from_trainer
datasets:
- gokuls/wiki_book_corpus_complete_processed_bert_dataset
metrics:
- accuracy
model-index:
- name: HBERTv1_emb_compress_48_L10_H64_A2
  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: 0.12810067638829456
---

<!-- 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_L10_H64_A2

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: 6.4110
- Accuracy: 0.1281

## 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: 96
- eval_batch_size: 96
- 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 |
|:-------------:|:-----:|:------:|:---------------:|:--------:|
| 8.6666        | 0.16  | 10000  | 8.5962          | 0.0465   |
| 7.2486        | 0.33  | 20000  | 7.2447          | 0.0467   |
| 7.0176        | 0.49  | 30000  | 7.0089          | 0.0670   |
| 6.8859        | 0.66  | 40000  | 6.8795          | 0.0840   |
| 6.7911        | 0.82  | 50000  | 6.7857          | 0.0918   |
| 6.7203        | 0.98  | 60000  | 6.7210          | 0.0952   |
| 6.6722        | 1.15  | 70000  | 6.6715          | 0.1004   |
| 6.6362        | 1.31  | 80000  | 6.6338          | 0.1033   |
| 6.5995        | 1.47  | 90000  | 6.6017          | 0.1065   |
| 6.5756        | 1.64  | 100000 | 6.5755          | 0.1092   |
| 6.5546        | 1.8   | 110000 | 6.5521          | 0.1118   |
| 6.5302        | 1.97  | 120000 | 6.5330          | 0.1138   |
| 6.5121        | 2.13  | 130000 | 6.5166          | 0.1156   |
| 6.5049        | 2.29  | 140000 | 6.5005          | 0.1174   |
| 6.483         | 2.46  | 150000 | 6.4869          | 0.1190   |
| 6.4757        | 2.62  | 160000 | 6.4755          | 0.1205   |
| 6.4659        | 2.79  | 170000 | 6.4655          | 0.1219   |
| 6.4527        | 2.95  | 180000 | 6.4569          | 0.1227   |
| 6.4517        | 3.11  | 190000 | 6.4477          | 0.1237   |
| 6.441         | 3.28  | 200000 | 6.4422          | 0.1245   |
| 6.4385        | 3.44  | 210000 | 6.4353          | 0.1253   |
| 6.4308        | 3.6   | 220000 | 6.4295          | 0.1256   |
| 6.4188        | 3.77  | 230000 | 6.4250          | 0.1264   |
| 6.422         | 3.93  | 240000 | 6.4213          | 0.1269   |
| 6.416         | 4.1   | 250000 | 6.4180          | 0.1273   |
| 6.4215        | 4.26  | 260000 | 6.4151          | 0.1276   |
| 6.4135        | 4.42  | 270000 | 6.4142          | 0.1277   |
| 6.4138        | 4.59  | 280000 | 6.4118          | 0.1281   |
| 6.41          | 4.75  | 290000 | 6.4097          | 0.1283   |
| 6.4114        | 4.92  | 300000 | 6.4103          | 0.1281   |


### Framework versions

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