<!-- 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. -->
flan-t5-base-codesearchnet-python3
This model is a fine-tuned version of google/flan-t5-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0641
- Rouge1: 0.6352
- Rouge2: 0.6162
- Avg Length: 16.995
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 | 23.1723 | 0.226 | 0.1296 | 18.7624 |
| No log | 2.0 | 78 | 2.9425 | 0.0041 | 0.0022 | 0.5596 |
| No log | 3.0 | 117 | 0.2351 | 0.0 | 0.0 | 0.0016 |
| No log | 3.99 | 156 | 0.0719 | 0.6245 | 0.6058 | 17.041 |
| No log | 4.99 | 195 | 0.0641 | 0.6352 | 0.6162 | 16.995 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
