Update main.py
Browse files
main.py
CHANGED
@@ -1,10 +1,8 @@
|
|
1 |
-
from fastapi import FastAPI,
|
2 |
-
from fastapi.responses import JSONResponse, FileResponse
|
3 |
from fastapi.staticfiles import StaticFiles
|
4 |
-
from fastapi.templating import Jinja2Templates
|
5 |
-
from fastapi import Request, Form
|
6 |
from huggingface_hub import InferenceClient
|
7 |
-
import
|
8 |
|
9 |
app = FastAPI()
|
10 |
|
@@ -23,7 +21,7 @@ MAX_TOKENS = 2000
|
|
23 |
TEMPERATURE = 0.7
|
24 |
TOP_P = 0.95
|
25 |
|
26 |
-
def respond(message, history: list[tuple[str, str]]):
|
27 |
messages = [{"role": "system", "content": SYSTEM_MESSAGE}]
|
28 |
|
29 |
for val in history:
|
@@ -51,12 +49,12 @@ def respond(message, history: list[tuple[str, str]]):
|
|
51 |
async def generate(request: Request):
|
52 |
form = await request.form()
|
53 |
prompt = form.get("prompt")
|
54 |
-
history = form.get("history", "[]") # Default to empty history
|
55 |
|
56 |
if not prompt:
|
57 |
raise HTTPException(status_code=400, detail="Prompt is required")
|
58 |
|
59 |
-
response_generator = respond(prompt,
|
60 |
final_response = ""
|
61 |
async for part in response_generator:
|
62 |
final_response += part
|
|
|
1 |
+
from fastapi import FastAPI, Request, HTTPException
|
2 |
+
from fastapi.responses import JSONResponse, FileResponse
|
3 |
from fastapi.staticfiles import StaticFiles
|
|
|
|
|
4 |
from huggingface_hub import InferenceClient
|
5 |
+
import json
|
6 |
|
7 |
app = FastAPI()
|
8 |
|
|
|
21 |
TEMPERATURE = 0.7
|
22 |
TOP_P = 0.95
|
23 |
|
24 |
+
async def respond(message, history: list[tuple[str, str]]):
|
25 |
messages = [{"role": "system", "content": SYSTEM_MESSAGE}]
|
26 |
|
27 |
for val in history:
|
|
|
49 |
async def generate(request: Request):
|
50 |
form = await request.form()
|
51 |
prompt = form.get("prompt")
|
52 |
+
history = json.loads(form.get("history", "[]")) # Default to empty history
|
53 |
|
54 |
if not prompt:
|
55 |
raise HTTPException(status_code=400, detail="Prompt is required")
|
56 |
|
57 |
+
response_generator = respond(prompt, history)
|
58 |
final_response = ""
|
59 |
async for part in response_generator:
|
60 |
final_response += part
|