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: 0.9051
- Bleu: 0.0187
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: 8
- 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: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu |
|---|---|---|---|---|
| 1.111 | 1.0 | 1002 | 0.9781 | 0.0108 |
| 0.998 | 2.0 | 2004 | 0.9397 | 0.0109 |
| 0.9295 | 3.0 | 3006 | 0.9204 | 0.0120 |
| 0.8814 | 4.0 | 4008 | 0.9088 | 0.0159 |
| 0.8557 | 5.0 | 5010 | 0.9064 | 0.0171 |
| 0.8364 | 6.0 | 6012 | 0.9055 | 0.0180 |
| 0.8184 | 6.9933 | 7007 | 0.9051 | 0.0187 |
Framework versions
- Transformers 4.47.0
- Pytorch 2.5.1+cu121
- Datasets 3.3.1
- Tokenizers 0.21.0
- Downloads last month
- 10
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for softhell/code_docstring_model
Base model
Salesforce/codet5-baseDataset used to train softhell/code_docstring_model
Evaluation results
- Bleu on code_search_netvalidation set self-reported0.019
# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("softhell/code_docstring_model") model = AutoModelForSeq2SeqLM.from_pretrained("softhell/code_docstring_model")