coedit-base
This model is a fine-tuned version of google/flan-t5-base on the CoEdIT dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5978
- Rouge1: 60.5931
- Rouge2: 48.0165
- Rougel: 57.8997
- Rougelsum: 57.9335
- Gen Len: 16.6729
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: 1e-05
- train_batch_size: 10
- eval_batch_size: 10
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 0.7478 | 1.0 | 6908 | 0.6452 | 59.7569 | 46.3099 | 56.4301 | 56.4464 | 16.6268 |
| 0.7127 | 2.0 | 13816 | 0.6086 | 60.2082 | 47.27 | 57.2356 | 57.2531 | 16.6513 |
| 0.7136 | 3.0 | 20724 | 0.6059 | 60.3747 | 47.6257 | 57.595 | 57.6184 | 16.6349 |
| 0.7038 | 4.0 | 27632 | 0.5999 | 60.5075 | 47.7856 | 57.7316 | 57.7698 | 16.6735 |
| 0.6911 | 5.0 | 34540 | 0.5978 | 60.5931 | 48.0165 | 57.8997 | 57.9335 | 16.6729 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.14.7
- Tokenizers 0.15.0