README.md

metadata

title: Espa AI - Code Assistant
emoji: 🚀
colorFrom: blue
colorTo: indigo
sdk: gradio
sdk_version: 5.18.0
app_file: app.py
pinned: false

Espa AI - Code Assistant

Espa AI is a fine-tuned version of CodeLlama-7B with LoRA adapters, designed to assist with coding tasks. It can generate code snippets, debug code, and answer programming-related questions.

Usage

Using the Hugging Face API

from transformers import AutoTokenizer, AutoModelForCausalLM

model_name = "DilipKY/esp-ai-lora"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

input_text = "Write Python code to reverse a string"
inputs = tokenizer(input_text, return_tensors="pt")
outputs = model.generate(**inputs, max_length=200)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))