File size: 2,102 Bytes
d6ae916 | 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 66 67 68 69 70 71 72 | # app.py
from fastapi import FastAPI, HTTPException
from pydantic import BaseModel
from typing import List, Tuple, Dict
import helper
import dialogue
import models.linear as linear_model
import models.logistic as logistic_model
import judge
app = FastAPI(
title="KindCare Chat API with ML",
description="Chat + prediction + judge endpoints"
)
class ChatRequest(BaseModel):
message: str
history: List[Tuple[str, str]] = []
class ChatResponse(BaseModel):
reply: str
history: List[Tuple[str, str]]
class PredictRequest(BaseModel):
features: Dict[str, float]
class PredictLinearResponse(BaseModel):
prediction: float
metrics: Dict[str, float]
class PredictLogisticRequest(PredictRequest):
bins: List[str]
class PredictLogisticResponse(BaseModel):
prediction: str
metrics: Dict[str, float]
class JudgeRequest(BaseModel):
model_type: str
metrics: Dict[str, float]
class JudgeResponse(BaseModel):
verdict: str
comments: List[str]
@app.post("/chat", response_model=ChatResponse)
async def chat(req: ChatRequest):
try:
reply, hist = dialogue.handle_message(req.message, req.history)
return ChatResponse(reply=reply, history=hist)
except Exception as e:
raise HTTPException(status_code=500, detail=str(e))
@app.post("/predict/linear", response_model=PredictLinearResponse)
async def predict_linear(req: PredictRequest):
pred, metrics = linear_model.predict(req.features)
return PredictLinearResponse(prediction=pred, metrics=metrics)
@app.post("/predict/logistic", response_model=PredictLogisticResponse)
async def predict_logistic(req: PredictLogisticRequest):
pred, metrics = logistic_model.predict(req.features, req.bins)
return PredictLogisticResponse(prediction=pred, metrics=metrics)
@app.post("/judge/model", response_model=JudgeResponse)
async def judge_endpoint(req: JudgeRequest):
result = judge.judge_model(req.model_type, req.metrics)
return JudgeResponse(**result)
if __name__ == "__main__":
import uvicorn
uvicorn.run("app:app", host="0.0.0.0", port=8000)
|