<!-- 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. -->
mt5-small-codesearchnet-python3
This model is a fine-tuned version of google/mt5-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 13.1295
- Rouge1: 0.0367
- Rouge2: 0.0116
- Avg Length: 17.0986
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 256
- 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 | Avg Length |
|---|---|---|---|---|---|---|
| No log | 1.0 | 39 | 49.9618 | 0.0214 | 0.0051 | 4.6036 |
| No log | 2.0 | 78 | 43.5835 | 0.0341 | 0.0112 | 6.5946 |
| No log | 3.0 | 117 | 33.6272 | 0.0633 | 0.0288 | 11.0158 |
| No log | 3.99 | 156 | 23.0445 | 0.0899 | 0.0444 | 13.4518 |
| No log | 4.99 | 195 | 13.1295 | 0.0367 | 0.0116 | 17.0986 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
