metadata
license: bigcode-openrail-m
base_model: bigcode/starcoder
tags:
- generated_from_trainer
model-index:
- name: starcoder-tune-cpp2py-snippet1
results: []
starcoder-tune-cpp2py-snippet1
This model is a fine-tuned version of bigcode/starcoder on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3914
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: 0.0003
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 50
- training_steps: 500
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.7375 | 0.1 | 50 | 0.3869 |
| 0.4809 | 0.2 | 100 | 0.3847 |
| 0.4753 | 0.3 | 150 | 0.3898 |
| 0.4472 | 0.4 | 200 | 0.3882 |
| 0.4509 | 0.5 | 250 | 0.3938 |
| 0.4369 | 0.6 | 300 | 0.3897 |
| 0.432 | 0.7 | 350 | 0.3968 |
| 0.4324 | 0.8 | 400 | 0.3917 |
| 0.4219 | 0.9 | 450 | 0.3920 |
| 0.4314 | 1.0 | 500 | 0.3914 |
Framework versions
- Transformers 4.32.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3