mt5-summarize-sum-test-internal
This model is a fine-tuned version of raquelclemente/mt5-summarize-sum on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0371
- Rouge1: 0.4910
- Rouge2: 0.3100
- Rougel: 0.3734
- Rougelsum: 0.3734
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: 0.0005
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 19
- num_epochs: 19
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
---|---|---|---|---|---|---|---|
No log | 1.9 | 5 | 1.0002 | 0.4672 | 0.3016 | 0.3936 | 0.3936 |
0.9822 | 3.81 | 10 | 0.9942 | 0.3989 | 0.3127 | 0.3713 | 0.3713 |
0.9822 | 5.71 | 15 | 1.0433 | 0.5093 | 0.3425 | 0.4298 | 0.4298 |
0.8459 | 7.62 | 20 | 1.0712 | 0.5394 | 0.3463 | 0.4320 | 0.4320 |
0.8459 | 9.52 | 25 | 1.0358 | 0.4281 | 0.2476 | 0.3160 | 0.3160 |
0.642 | 11.43 | 30 | 1.0335 | 0.4439 | 0.2582 | 0.3224 | 0.3224 |
0.642 | 13.33 | 35 | 1.0371 | 0.4910 | 0.3100 | 0.3734 | 0.3734 |
Framework versions
- Transformers 4.27.4
- Pytorch 1.13.0
- Datasets 2.1.0
- Tokenizers 0.13.2
- Downloads last month
- 1