metadata

library_name: transformers
license: apache-2.0
base_model: Salesforce/codet5-base
tags:
  - generated_from_trainer
datasets:
  - code_search_net
metrics:
  - bleu
model-index:
  - name: code_docstring_model
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: code_search_net
          type: code_search_net
          config: python
          split: validation
          args: python
        metrics:
          - name: Bleu
            type: bleu
            value: 0.013233021060148939

code_docstring_model

This model is a fine-tuned version of Salesforce/codet5-base on the code_search_net dataset. It achieves the following results on the evaluation set:

Loss: 0.9120
Bleu: 0.0132

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: 4
total_train_batch_size: 64
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
1.0906	1.0	1002	0.9715	0.0107
0.9922	2.0	2004	0.9390	0.0108
0.9325	3.0	3006	0.9233	0.0113
0.8936	4.0	4008	0.9134	0.0124
0.8769	4.9953	5005	0.9120	0.0132

Framework versions

Transformers 4.47.0
Pytorch 2.5.1+cu121
Datasets 3.3.1
Tokenizers 0.21.0