Spaces:
Sleeping
Sleeping
File size: 1,752 Bytes
7a8c6cb 7752684 7a8c6cb 7752684 7a8c6cb 7752684 7a8c6cb 7752684 1984220 7a8c6cb 7752684 7a8c6cb 7752684 7a8c6cb 7752684 7a8c6cb 7752684 7a8c6cb 7752684 7a8c6cb 7752684 7a8c6cb 7752684 |
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 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 |
from langchain_community.llms.ctransformers import CTransformers
from langchain.chains.llm import LLMChain
from langchain.prompts import PromptTemplate
import os
import gradio as gr
import time
custom_prompt_template = """
You are an AI coding assistant and your task is to solve coding problems
and return code snippets based on the user's query. Below is the user's query.
Query: {query}
You just return the helpful code and related details.
Helpful code and related details:
"""
def set_custom_prompt():
prompt = PromptTemplate(
template=custom_prompt_template,
input_variables=['query']
)
return prompt
def load_model():
llm = CTransformers(
model='TheBloke/CodeLlama-7B-Instruct-GGML',
model_type='llama',
max_new_tokens=1096,
temperature=0.2,
repetition_penalty=1.13
)
return llm
def chain_pipeline():
llm = load_model()
qa_prompt = set_custom_prompt()
qa_chain = LLMChain(
prompt=qa_prompt,
llm=llm
)
return qa_chain
llmchain = chain_pipeline()
def bot(query):
llm_response = llmchain.run({'query': query})
# Wrap the response in triple backticks for code formatting
formatted_response = f"```\n{llm_response}\n```"
return formatted_response
with gr.Blocks(title="Can AI code ?") as demo:
gr.Markdown('# Code LLAMA demo')
chatbot = gr.Chatbot([], elem_id='chatbot', height=700)
msg = gr.Textbox()
clear = gr.ClearButton([msg, chatbot])
def respond(message, chat_history):
bot_message = bot(message)
chat_history.append((message, bot_message))
time.sleep(2)
return "", chat_history
msg.submit(respond, [msg, chatbot], [msg, chatbot])
demo.launch()
|