File size: 1,727 Bytes
3db2b26
 
 
2c1c3e8
 
3db2b26
 
 
 
 
 
 
 
 
 
 
2c1c3e8
3c65d1a
 
 
 
3db2b26
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3c65d1a
2c1c3e8
3db2b26
 
 
2c1c3e8
3c65d1a
3db2b26
 
 
3c65d1a
 
 
 
 
 
 
3db2b26
 
 
 
2c1c3e8
3db2b26
 
 
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
---
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: BERT2BERT_pretrained_LC-QuAD_2.0
  results: []
---

<!-- 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. -->

# BERT2BERT_pretrained_LC-QuAD_2.0

This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0571
- Bleu: 75.7356
- Em: 0.0432
- Gen Len: 47.2896

## 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: 5e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2500
- num_epochs: 8.0

### Training results

| Training Loss | Epoch | Step  | Bleu    | Em     | Gen Len  | Validation Loss |
|:-------------:|:-----:|:-----:|:-------:|:------:|:--------:|:---------------:|
| 6.6044        | 1.4   | 2000  | 3.8045  | 0.0    | 158.8473 | 6.1713          |
| 3.032         | 2.8   | 4000  | 27.701  | 0.0    | 33.9568  | 2.9045          |
| 1.5893        | 4.19  | 6000  | 57.7509 | 0.0033 | 44.132   | 1.7327          |
| 1.0512        | 5.59  | 8000  | 71.1272 | 0.0241 | 46.0672  | 1.2382          |
| 0.7515        | 6.99  | 10000 | 1.0571  | 75.7356| 0.0432   | 47.2896         |


### Framework versions

- Transformers 4.30.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3