Instructions to use andstor/Qwen-Qwen2.5-Coder-7B-unit-test-prompt-tuning with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use andstor/Qwen-Qwen2.5-Coder-7B-unit-test-prompt-tuning with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-Coder-7B") model = PeftModel.from_pretrained(base_model, "andstor/Qwen-Qwen2.5-Coder-7B-unit-test-prompt-tuning") - Notebooks
- Google Colab
- Kaggle
Download eval_results.json from andstor/Qwen-Qwen2.5-Coder-7B-unit-test-prompt-tuning: direct link, hf CLI and curl.
- Browser
- Download file 278 Bytes
-
https://huggingface.co/andstor/Qwen-Qwen2.5-Coder-7B-unit-test-prompt-tuning/resolve/6c82a5ff9f3d892de66dd6a265bfe72aa59d016d/eval_results.json
- Command line
-
hf download hf://andstor/Qwen-Qwen2.5-Coder-7B-unit-test-prompt-tuning@6c82a5ff9f3d892de66dd6a265bfe72aa59d016d/eval_results.json
-
curl -L -o eval_results.json https://huggingface.co/andstor/Qwen-Qwen2.5-Coder-7B-unit-test-prompt-tuning/resolve/6c82a5ff9f3d892de66dd6a265bfe72aa59d016d/eval_results.json
278 Bytes
| { | |
| "epoch": 2.999179655455291, | |
| "eval_accuracy": 0.733010457290466, | |
| "eval_loss": 0.7598717212677002, | |
| "eval_runtime": 25.7382, | |
| "eval_samples": 934, | |
| "eval_samples_per_second": 36.288, | |
| "eval_steps_per_second": 18.144, | |
| "perplexity": 2.138001942726325 | |
| } |