<!-- 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. -->
my_finetuned_t5_cnn_model
This model is a fine-tuned version of t5-small on the cnn_dailymail dataset. It achieves the following results on the evaluation set:
- Loss: 1.9870
- Rouge1: 0.2484
- Rouge2: 0.1
- Rougel: 0.2024
- Rougelsum: 0.2025
- Gen Len: 19.0
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
---|---|---|---|---|---|---|---|---|
2.0391 | 1.0 | 657 | 1.9948 | 0.2479 | 0.0986 | 0.2013 | 0.2012 | 19.0 |
1.9686 | 2.0 | 1314 | 1.9895 | 0.2476 | 0.0985 | 0.2016 | 0.2013 | 19.0 |
1.9495 | 3.0 | 1971 | 1.9883 | 0.2482 | 0.1001 | 0.202 | 0.202 | 19.0 |
1.9394 | 4.0 | 2628 | 1.9870 | 0.2484 | 0.1 | 0.2024 | 0.2025 | 19.0 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3