gpt4free/interference/app.py

161 lines
4.9 KiB
Python
Raw Normal View History

2023-06-24 04:47:00 +03:00
import json
import random
2023-07-28 13:07:17 +03:00
import string
import time
from typing import Any
import requests
2023-07-28 13:07:17 +03:00
from flask import Flask, request
2023-06-24 04:47:00 +03:00
from flask_cors import CORS
from transformers import AutoTokenizer
2023-07-28 13:07:17 +03:00
from g4f import ChatCompletion
2023-06-24 04:47:00 +03:00
app = Flask(__name__)
CORS(app)
2023-08-31 14:32:23 +03:00
2023-07-28 13:07:17 +03:00
@app.route("/chat/completions", methods=["POST"])
2023-06-24 04:47:00 +03:00
def chat_completions():
2023-07-28 13:07:17 +03:00
model = request.get_json().get("model", "gpt-3.5-turbo")
stream = request.get_json().get("stream", False)
messages = request.get_json().get("messages")
2023-06-24 04:47:00 +03:00
2023-07-28 13:07:17 +03:00
response = ChatCompletion.create(model=model, stream=stream, messages=messages)
2023-06-24 04:47:00 +03:00
2023-07-28 13:07:17 +03:00
completion_id = "".join(random.choices(string.ascii_letters + string.digits, k=28))
completion_timestamp = int(time.time())
if not stream:
2023-06-24 04:47:00 +03:00
return {
2023-07-28 13:07:17 +03:00
"id": f"chatcmpl-{completion_id}",
"object": "chat.completion",
"created": completion_timestamp,
"model": model,
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": response,
},
"finish_reason": "stop",
}
],
"usage": {
"prompt_tokens": None,
"completion_tokens": None,
"total_tokens": None,
2023-06-24 04:47:00 +03:00
},
}
2023-07-28 13:07:17 +03:00
def streaming():
for chunk in response:
2023-06-24 04:47:00 +03:00
completion_data = {
2023-07-28 13:07:17 +03:00
"id": f"chatcmpl-{completion_id}",
"object": "chat.completion.chunk",
"created": completion_timestamp,
"model": model,
"choices": [
2023-06-24 04:47:00 +03:00
{
2023-07-28 13:07:17 +03:00
"index": 0,
"delta": {
"content": chunk,
2023-06-24 04:47:00 +03:00
},
2023-07-28 13:07:17 +03:00
"finish_reason": None,
2023-06-24 04:47:00 +03:00
}
2023-07-28 13:07:17 +03:00
],
2023-06-24 04:47:00 +03:00
}
2023-07-28 13:07:17 +03:00
content = json.dumps(completion_data, separators=(",", ":"))
yield f"data: {content}\n\n"
2023-06-24 04:47:00 +03:00
time.sleep(0.1)
2023-07-28 13:07:17 +03:00
end_completion_data: dict[str, Any] = {
"id": f"chatcmpl-{completion_id}",
"object": "chat.completion.chunk",
"created": completion_timestamp,
"model": model,
"choices": [
{
"index": 0,
"delta": {},
"finish_reason": "stop",
}
],
}
content = json.dumps(end_completion_data, separators=(",", ":"))
yield f"data: {content}\n\n"
2023-06-24 04:47:00 +03:00
2023-07-28 13:07:17 +03:00
return app.response_class(streaming(), mimetype="text/event-stream")
2023-06-24 04:47:00 +03:00
#Get the embedding from huggingface
def get_embedding(input_text, token):
huggingface_token = token
embedding_model = "sentence-transformers/all-mpnet-base-v2"
max_token_length = 500
# Load the tokenizer for the "all-mpnet-base-v2" model
tokenizer = AutoTokenizer.from_pretrained(embedding_model)
# Tokenize the text and split the tokens into chunks of 500 tokens each
tokens = tokenizer.tokenize(input_text)
token_chunks = [tokens[i:i + max_token_length] for i in range(0, len(tokens), max_token_length)]
# Initialize an empty list
embeddings = []
# Create embeddings for each chunk
for chunk in token_chunks:
# Convert the chunk tokens back to text
chunk_text = tokenizer.convert_tokens_to_string(chunk)
# Use the Hugging Face API to get embeddings for the chunk
api_url = f"https://api-inference.huggingface.co/pipeline/feature-extraction/{embedding_model}"
headers = {"Authorization": f"Bearer {huggingface_token}"}
chunk_text = chunk_text.replace("\n", " ")
# Make a POST request to get the chunk's embedding
response = requests.post(api_url, headers=headers, json={"inputs": chunk_text, "options": {"wait_for_model": True}})
# Parse the response and extract the embedding
chunk_embedding = response.json()
# Append the embedding to the list
embeddings.append(chunk_embedding)
#averaging all the embeddings
#this isn't very effective
#someone a better idea?
num_embeddings = len(embeddings)
average_embedding = [sum(x) / num_embeddings for x in zip(*embeddings)]
embedding = average_embedding
return embedding
@app.route("/embeddings", methods=["POST"])
def embeddings():
input_text_list = request.get_json().get("input")
input_text = ' '.join(map(str, input_text_list))
token = request.headers.get('Authorization').replace("Bearer ", "")
embedding = get_embedding(input_text, token)
return {
"data": [
{
"embedding": embedding,
"index": 0,
"object": "embedding"
}
],
"model": "text-embedding-ada-002",
"object": "list",
"usage": {
"prompt_tokens": None,
"total_tokens": None
}
}
2023-08-31 14:32:23 +03:00
def main():
app.run(host="0.0.0.0", port=1337, debug=True)
2023-07-28 13:07:17 +03:00
if __name__ == "__main__":
main()