File size: 1,848 Bytes
3f9f94a
83b5a87
 
 
 
 
 
 
 
 
 
3f9f94a
 
83b5a87
 
3f9f94a
83b5a87
3f9f94a
83b5a87
 
 
 
 
 
3f9f94a
83b5a87
3f9f94a
83b5a87
3f9f94a
83b5a87
3f9f94a
83b5a87
3f9f94a
83b5a87
3f9f94a
83b5a87
3f9f94a
83b5a87
3f9f94a
83b5a87
3f9f94a
83b5a87
 
 
 
 
 
 
 
 
3f9f94a
83b5a87
3f9f94a
83b5a87
 
 
 
 
 
 
3f9f94a
 
83b5a87
3f9f94a
83b5a87
 
 
 
 
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
---
base_model: UBC-NLP/AraT5v2-base-1024
library_name: peft
metrics:
- bleu
- rouge
tags:
- generated_from_trainer
model-index:
- name: araT5-Base-with-LoRA
  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. -->

# araT5-Base-with-LoRA

This model is a fine-tuned version of [UBC-NLP/AraT5v2-base-1024](https://huggingface.co/UBC-NLP/AraT5v2-base-1024) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.1185
- Bleu: 12.5314
- Rouge: 0.5045
- Gen Len: 14.066

## 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: 0.0002
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Bleu    | Rouge  | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:------:|:-------:|
| 4.2656        | 1.0   | 7500  | 2.6346          | 9.6457  | 0.4263 | 14.0196 |
| 3.2885        | 2.0   | 15000 | 2.3718          | 10.7953 | 0.4683 | 14.1228 |
| 3.0225        | 3.0   | 22500 | 2.2546          | 11.4633 | 0.4853 | 14.0032 |
| 2.865         | 4.0   | 30000 | 2.1520          | 12.0114 | 0.4999 | 13.9988 |
| 2.747         | 5.0   | 37500 | 2.1185          | 12.5314 | 0.5045 | 14.066  |


### Framework versions

- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1