metadata
license: bsd-3-clause
base_model: Salesforce/codet5p-220m
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
model-index:
- name: Salesforce-codet5p-220m-finetuned-defect-cwe-group
results: []
Salesforce-codet5p-220m-finetuned-defect-cwe-group
This model is a fine-tuned version of Salesforce/codet5p-220m on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5618
- Accuracy: 0.7428
- Precision: 0.5937
- Recall: 0.4798
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: 8
- eval_batch_size: 8
- seed: 4711
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall |
|---|---|---|---|---|---|---|
| No log | 1.0 | 462 | 0.6991 | 0.6911 | 0.6402 | 0.3911 |
| 0.803 | 2.0 | 925 | 0.6093 | 0.7192 | 0.6387 | 0.4320 |
| 0.6422 | 3.0 | 1387 | 0.5770 | 0.7254 | 0.5693 | 0.4681 |
| 0.5365 | 4.0 | 1850 | 0.5672 | 0.7248 | 0.5682 | 0.4721 |
| 0.4489 | 4.99 | 2310 | 0.5618 | 0.7428 | 0.5937 | 0.4798 |
Framework versions
- Transformers 4.38.1
- Pytorch 2.2.1+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2