Gemma 2B Finetuned Python Model

🚀 Gemma-2B Fine-Tuned Python Model

A fine-tuned deep learning model based on Gemma-2B architecture for Python programming tasks, offering code suggestions and improvements.

🚀 Quick Start

This Gemma-2B fine-tuned model is tailored for Python programming. It can understand Python code and assist developers in various ways, such as providing code suggestions and completing snippets.

✨ Features

• Code Completion: Automatically finish code snippets from partial inputs.
• Syntax Correction: Spot and suggest fixes for Python syntax errors.
• Code Quality Enhancement: Offer advice to boost code readability, efficiency, and maintainability.
• Debugging Support: Provide insights to help debug Python code by identifying potential issues.

📦 Installation

1. Install Gemma Python Package:

    pip install -q -U transformers==4.38.0
    pip install torch
   

💻 Usage Examples

Basic Usage

# Load model directly
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("suriya7/Gemma-2B-Finetuned-Python-Model")
model = AutoModelForCausalLM.from_pretrained("suriya7/Gemma-2B-Finetuned-Python-Model")

query = input('enter a query:')
prompt_template = f"""
<start_of_turn>user based on given instruction create a solution\n\nhere are the instruction {query}
<end_of_turn>\n<start_of_turn>model
"""
prompt = prompt_template
encodeds = tokenizer(prompt, return_tensors="pt", add_special_tokens=True).input_ids

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model.to(device)
inputs = encodeds.to(device)


# Increase max_new_tokens if needed
generated_ids = model.generate(inputs, max_new_tokens=1000, do_sample=True, pad_token_id=tokenizer.eos_token_id)
ans = ''
for i in tokenizer.decode(generated_ids[0], skip_special_tokens=True).split('<end_of_turn>')[:2]:
    ans += i

# Extract only the model's answer
model_answer = ans.split("model")[1].strip()
print(model_answer)

📚 Documentation

Model Details

📄 License

This project is licensed under the MIT License.