File size: 2,108 Bytes
e0d28f4 |
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 |
---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- thaisum
metrics:
- rouge
model-index:
- name: mt5_thaisum_model
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: thaisum
type: thaisum
config: thaisum
split: validation
args: thaisum
metrics:
- name: Rouge1
type: rouge
value: 0.1432
---
<!-- 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_thaisum_model
This model is a fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) on the thaisum dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3540
- Rouge1: 0.1432
- Rouge2: 0.041
- Rougel: 0.1423
- Rougelsum: 0.142
- Gen Len: 18.933
## 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: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|
| 2.4271 | 1.0 | 2500 | 0.3976 | 0.1306 | 0.0372 | 0.1302 | 0.1299 | 18.9665 |
| 2.1992 | 2.0 | 5000 | 0.3720 | 0.1392 | 0.0376 | 0.1382 | 0.1384 | 18.922 |
| 2.1687 | 3.0 | 7500 | 0.3599 | 0.1401 | 0.0391 | 0.1394 | 0.1389 | 18.9215 |
| 2.1096 | 4.0 | 10000 | 0.3540 | 0.1432 | 0.041 | 0.1423 | 0.142 | 18.933 |
### Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
|