File size: 1,725 Bytes
2b1c96d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
tags:
- generated_from_trainer
model-index:
- name: mT5_multilingual_XLSum-marathi-summarization
  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. -->

# mT5_multilingual_XLSum-marathi-summarization

This model is a fine-tuned version of [csebuetnlp/mT5_multilingual_XLSum](https://huggingface.co/csebuetnlp/mT5_multilingual_XLSum) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0760

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

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 0.5994        | 0.55  | 100  | 0.1662          |
| 0.2766        | 1.09  | 200  | 0.1032          |
| 0.1967        | 1.64  | 300  | 0.0881          |
| 0.194         | 2.19  | 400  | 0.0850          |
| 0.1667        | 2.73  | 500  | 0.0821          |
| 0.1615        | 3.28  | 600  | 0.0780          |
| 0.1368        | 3.83  | 700  | 0.0783          |
| 0.1372        | 4.37  | 800  | 0.0762          |
| 0.1265        | 4.92  | 900  | 0.0760          |


### Framework versions

- Transformers 4.30.2
- Pytorch 2.0.0+cpu
- Datasets 2.1.0
- Tokenizers 0.13.3