File size: 1,270 Bytes
87ed98d 9702672 4b74937 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 |
import gradio as gr
from huggingface_hub import InferenceClient
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
# Base model (LLaMA 3.1 8B) from Meta
base_model_name = "meta-llama/Llama-3-8B"
# Your fine-tuned LoRA adapter (uploaded to Hugging Face)
lora_model_name = "starnernj/Early-Christian-Church-Fathers-LLaMA-3.1-Fine-Tuned"
# Load base model
model = AutoModelForCausalLM.from_pretrained(base_model_name)
# Load LoRA adapter
model = PeftModel.from_pretrained(model, lora_model_name)
# Load tokenizer
tokenizer = AutoTokenizer.from_pretrained(base_model_name)
# Function to generate responses
def chatbot_response(user_input):
inputs = tokenizer(user_input, return_tensors="pt")
outputs = model.generate(**inputs, max_length=400)
return tokenizer.decode(outputs[0], skip_special_tokens=True)
# Launch the Gradio chatbot
interface = gr.Interface(
fn=chatbot_response,
inputs=gr.Textbox(lines=2, placeholder="Ask me anything..."),
outputs="text",
title="Early Christian Church Fathers Fine-Tuned LLaMA 3.1 8B with LoRA",
description="A chatbot using my fine-tuned LoRA adapter on LLaMA 3.1 8B, tuned on thousands of writings of the early Christian Church Fathers.",
)
interface.launch() |