metadata
license: apache-2.0
tags:
- summarization
- generated_from_trainer
datasets:
- snli
metrics:
- rouge
model-index:
- name: t5-small-finetuned-contradiction
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: snli
type: snli
args: plain_text
metrics:
- name: Rouge1
type: rouge
value: 34.4713
t5-small-finetuned-contradiction
This model is a fine-tuned version of domenicrosati/t5-small-finetuned-contradiction on the snli dataset. It achieves the following results on the evaluation set:
- Loss: 2.0524
- Rouge1: 34.4713
- Rouge2: 14.6253
- Rougel: 32.5971
- Rougelsum: 32.5854
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| 1.9316 | 1.0 | 2863 | 2.0832 | 34.5611 | 14.5831 | 32.7994 | 32.7887 |
| 1.9898 | 2.0 | 5726 | 2.0742 | 34.4883 | 14.6727 | 32.6392 | 32.6204 |
| 1.9995 | 3.0 | 8589 | 2.0658 | 34.6327 | 14.6086 | 32.7722 | 32.7524 |
| 2.0108 | 4.0 | 11452 | 2.0602 | 34.6013 | 14.6843 | 32.7286 | 32.7192 |
| 2.0165 | 5.0 | 14315 | 2.0581 | 34.5423 | 14.6649 | 32.705 | 32.6891 |
| 2.0132 | 6.0 | 17178 | 2.0545 | 34.6902 | 14.7817 | 32.8538 | 32.8374 |
| 2.0128 | 7.0 | 20041 | 2.0537 | 34.5965 | 14.691 | 32.7323 | 32.7165 |
| 2.0139 | 8.0 | 22904 | 2.0524 | 34.529 | 14.6524 | 32.6635 | 32.6443 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.11.0+cu102
- Datasets 2.1.0
- Tokenizers 0.12.1