metadata
license: apache-2.0
base_model: google/mt5-small
tags:
- summarization
- generated_from_trainer
datasets:
- gazeta
metrics:
- rouge
model-index:
- name: mt5-small-finetuned-gazeta-ru
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: gazeta
type: gazeta
config: default
split: validation
args: default
metrics:
- name: Rouge1
type: rouge
value: 2.9422
mt5-small-finetuned-gazeta-ru
This model is a fine-tuned version of google/mt5-small on the gazeta dataset. It achieves the following results on the evaluation set:
- Loss: 3.3287
- Rouge1: 2.9422
- Rouge2: 0.25
- Rougel: 2.9053
- Rougelsum: 2.9131
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: 5.6e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| 5.5708 | 1.0 | 1690 | 3.3106 | 1.8563 | 0.1911 | 1.8332 | 1.8348 |
| 4.0219 | 2.0 | 3380 | 3.3048 | 2.2018 | 0.1649 | 2.1978 | 2.2022 |
| 3.7276 | 3.0 | 5070 | 3.3320 | 3.2293 | 0.2173 | 3.194 | 3.2039 |
| 3.5835 | 4.0 | 6760 | 3.3308 | 3.2189 | 0.2932 | 3.1825 | 3.1841 |
| 3.4944 | 5.0 | 8450 | 3.3104 | 2.8833 | 0.1964 | 2.8521 | 2.8537 |
| 3.4203 | 6.0 | 10140 | 3.3032 | 2.9914 | 0.2723 | 2.9516 | 2.9542 |
| 3.3774 | 7.0 | 11830 | 3.3232 | 2.9982 | 0.3063 | 2.965 | 2.9642 |
| 3.348 | 8.0 | 13520 | 3.3287 | 2.9422 | 0.25 | 2.9053 | 2.9131 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1