metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- null
metrics:
- rouge
model-index:
- name: mt5-small-finetuned-amazon-en-es
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
metrics:
- name: Rouge1
type: rouge
value: 8.8272
mt5-small-finetuned-amazon-en-es
This model is a fine-tuned version of google/mt5-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.3342
- Rouge1: 8.8272
- Rouge2: 2.5114
- Rougel: 8.6749
- Rougelsum: 8.6722
- Gen Len: 4.2877
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 9.4562 | 1.0 | 2202 | 3.5591 | 6.6009 | 1.7239 | 6.5036 | 6.5228 | 3.4434 |
| 4.6481 | 2.0 | 4404 | 3.3600 | 7.3535 | 1.9174 | 7.2846 | 7.3053 | 3.809 |
| 4.3333 | 3.0 | 6606 | 3.3342 | 8.8272 | 2.5114 | 8.6749 | 8.6722 | 4.2877 |
Framework versions
- Transformers 4.10.3
- Pytorch 1.9.1+cu102
- Datasets 1.12.1
- Tokenizers 0.10.3