metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- xlsum
metrics:
- rouge
model-index:
- name: mt5-swatf
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: xlsum
type: xlsum
config: swahili
split: test
args: swahili
metrics:
- name: Rouge1
type: rouge
value: 9.6904
mt5-swatf
This model is a fine-tuned version of google/mt5-small on the xlsum dataset. It achieves the following results on the evaluation set:
- Loss: nan
- Rouge1: 9.6904
- Rouge2: 1.3302
- Rougel: 8.4948
- Rougelsum: 8.497
- Gen Len: 685.8156
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: 8
- eval_batch_size: 4
- 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 | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 188 | nan | 9.6904 | 1.3302 | 8.4948 | 8.497 | 685.8156 |
| No log | 2.0 | 376 | nan | 9.6904 | 1.3302 | 8.4948 | 8.497 | 685.8156 |
| 0.0 | 3.0 | 564 | nan | 9.6904 | 1.3302 | 8.4948 | 8.497 | 685.8156 |
| 0.0 | 4.0 | 752 | nan | 9.6904 | 1.3302 | 8.4948 | 8.497 | 685.8156 |
| 0.0 | 5.0 | 940 | nan | 9.6904 | 1.3302 | 8.4948 | 8.497 | 685.8156 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3