<!-- 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. -->
starcoder_loraprompt_finetuned_cpp2py
HuggingFace's weights only contains the Prompt Tuning part. LoRA weights are not uploaded. This model card is for info only. This model is a fine-tuned version of bigcode/starcoder on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3771
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: 4e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 5
- training_steps: 600
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.8971 | 0.04 | 25 | 0.4637 |
| 0.5934 | 0.08 | 50 | 0.4148 |
| 0.5442 | 0.12 | 75 | 0.3877 |
| 0.5217 | 0.17 | 100 | 0.3799 |
| 0.5072 | 0.21 | 125 | 0.3820 |
| 0.4792 | 0.25 | 150 | 0.3785 |
| 0.4652 | 0.29 | 175 | 0.3813 |
| 0.4649 | 0.33 | 200 | 0.3819 |
| 0.4671 | 0.38 | 225 | 0.3822 |
| 0.477 | 0.42 | 250 | 0.3855 |
| 0.4829 | 0.46 | 275 | 0.3801 |
| 0.4638 | 0.5 | 300 | 0.3776 |
| 0.4678 | 0.54 | 325 | 0.3811 |
| 0.4952 | 0.58 | 350 | 0.3770 |
| 0.4755 | 0.62 | 375 | 0.3768 |
| 0.4622 | 0.67 | 400 | 0.3787 |
| 0.4656 | 0.71 | 425 | 0.3792 |
| 0.4734 | 0.75 | 450 | 0.3778 |
| 0.4737 | 0.79 | 475 | 0.3776 |
| 0.4714 | 0.83 | 500 | 0.3778 |
| 0.4451 | 0.88 | 525 | 0.3781 |
| 0.4457 | 0.92 | 550 | 0.3782 |
| 0.4465 | 0.96 | 575 | 0.3774 |
| 0.4541 | 1.0 | 600 | 0.3771 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.4
- Tokenizers 0.13.3
