Files
gpt-pilot/euclid/utils/llm_connection.py
Zvonimir Sabljic cb579d8aba Added local files
2023-07-18 11:17:22 +02:00

99 lines
3.2 KiB
Python

# llm_connection.py
import re
import requests
from dotenv import load_dotenv
import os
from tiktoken import Tokenizer
from typing import List
from http.server import BaseHTTPRequestHandler
from socketserver import ThreadingMixIn
from http.server import HTTPServer
from euclid.const.llm import MIN_TOKENS_FOR_GPT_RESPONSE, MAX_GPT_MODEL_TOKENS
from euclid.const.prompts import SYS_MESSAGE
from jinja2 import Environment, FileSystemLoader
def connect_to_llm():
pass
def get_user_flows(description):
prompt = get_prompt('breakdown_1_user_flows.prompt', {'description': description})
messages = [
SYS_MESSAGE['tdd_engineer'],
# app type
#
{"role": "user", "content": prompt},
]
create_gpt_chat_completion(messages, min_tokens=MIN_TOKENS_FOR_GPT_RESPONSE)
def get_prompt(prompt_name, data):
# Create a file system loader with the directory of the templates
file_loader = FileSystemLoader('../prompts')
# Create the Jinja2 environment
env = Environment(loader=file_loader)
# Load the template
template = env.get_template(prompt_name)
# Render the template with the provided data
output = template.render(data)
return output
def get_tokens_in_messages(messages: List[str]) -> int:
tokenizer = Tokenizer()
tokenized_messages = [tokenizer.encode(message) for message in messages]
return sum(len(tokens) for tokens in tokenized_messages)
def create_gpt_chat_completion(messages: List[dict], min_tokens=MIN_TOKENS_FOR_GPT_RESPONSE):
api_key = os.getenv("OPENAI_API_KEY")
tokens_in_messages = get_tokens_in_messages(messages)
if tokens_in_messages + min_tokens > MAX_GPT_MODEL_TOKENS:
raise ValueError(f'Too many tokens in messages: {tokens_in_messages}. Please try a different test.')
gpt_data = {
'model': 'gpt-4',
'n': 1,
'max_tokens': min(4096, MAX_GPT_MODEL_TOKENS - tokens_in_messages),
'messages': messages,
'stream': True
}
try:
return stream_gpt_completion(gpt_data, api_key)
except Exception as e:
print('The request to OpenAI API failed. Might be due to GPT being down or due to the too large message. It\'s best if you try another export.')
print(e)
def stream_gpt_completion(data, api_key):
response = requests.post(
'https://api.openai.com/v1/chat/completions',
headers={'Content-Type': 'application/json', 'Authorization': 'Bearer ' + api_key},
json=data,
stream=True
)
if response.status_code != 200:
print(f'problem with request: {response.text}')
return
gpt_response = ''
for line in response.iter_lines():
if line: # filter out keep-alive new lines
json_line = json.loads(line)
if 'error' in json_line or 'message' in json_line:
print(json_line, end="")
return
content = json_line.get('choices')[0]['message']['content']
gpt_response += content
print(content, end="")
new_code = postprocessing(gpt_response, 'user_flows') # TODO add type dynamically
return new_code
def postprocessing(gpt_response, type):
pass