README.md

---
license: apache-2.0
datasets:
- sahil2801/CodeAlpaca-20k
language:
- en
base_model:
- Qwen/Qwen2.5-Coder-0.5B
tags:
- code
---
# Qwen 2.5 Coder (0.5B) - C++ QA Fine-Tuned
This is a fine-tuned version of the highly capable [Qwen2.5-Coder-0.5B](https://huggingface.co/Qwen/Qwen2.5-Coder-0.5B) model. It has been specifically instruction-tuned to answer programming questions and write code for the **C++** programming language.
## Model Details
- **Base Model**: Qwen/Qwen2.5-Coder-0.5B
- **Parameters**: 500 Million
- **Language**: English / C++
- **Intended Use**: Answering C++ programming questions, generating C++ snippets, and code explanation.
## Training Data
This model was fine-tuned using a filtered subset of the `sahil2801/CodeAlpaca-20k` dataset. The training data was specifically filtered to only include instructions and inputs that reference `C++` or `cpp`, ensuring the model focuses heavily on this language domain.

## Fine-tuned parameters 
The model was fine-tuned using the Hugging Face trl library (SFTTrainer) with the following hyperparameters:

Optimizer: AdamW
Learning Rate: 2e-5
Batch Size: 1 (with gradient accumulation of 4)
Precision: fp16 (Mixed Precision)

## How to use
You can load and use this model directly with the `transformers` library:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model_id = "VesileHan/Qwen2.5_coder_cpp"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype=torch.float16
)
question = "How do I reverse a string in C++?"
prompt = f"Below is an instruction that describes a coding task. Write a response that appropriately completes the request.\n\n### Instruction:\n{question}\n\n### Response:\n"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=150, temperature=0.7)
print(tokenizer.decode(outputs[0][inputs['input_ids'].shape[1]:], skip_special_tokens=True))