Spaces:
Running
Running
File size: 8,919 Bytes
f9ce179 edd091c 36a0796 f9ce179 0d22171 f9ce179 72d7cd5 f9ce179 72d7cd5 |
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 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 |
# from fastapi import FastAPI, HTTPException, status, Depends
# from fastapi.responses import RedirectResponse
# from pydantic import BaseModel, conlist
# import pandas as pd
# from pycaret.classification import load_model, predict_model
# import logging
# from typing import Optional
# import numpy as np
# import os
# # Constants
# MODEL_PATH = "./api/model/saved_tuned_model" # os.getenv("MODEL_PATH", "saved_tuned_model") # Load model path from environment variable
# EMBEDDING_DIMENSION = 1024 # Update this to match your model's expected input dimension
# LOG_LEVEL = os.getenv("LOG_LEVEL", "INFO") # Logging level from environment variable
# # Configure logging
# logging.basicConfig(level=LOG_LEVEL)
# logger = logging.getLogger(__name__)
# # Load the saved model
# def load_tuned_model(model_path: str):
# """Load the pre-trained model from the specified path."""
# try:
# logger.info(f"Loading model from {model_path}...")
# model = load_model(model_path)
# logger.info("Model loaded successfully.")
# return model
# except Exception as e:
# logger.error(f"Failed to load the model: {str(e)}")
# raise RuntimeError(f"Model loading failed: {str(e)}")
# tuned_model = load_tuned_model(MODEL_PATH)
# # Define the input data model using Pydantic
# class EmbeddingRequest(BaseModel):
# embedding: conlist(
# float, min_length=EMBEDDING_DIMENSION, max_length=EMBEDDING_DIMENSION
# )
# # Define the response model
# class PredictionResponse(BaseModel):
# predicted_label: int
# predicted_score: float
# # Initialize FastAPI app
# app = FastAPI(
# title="Embedding Prediction API",
# description="API for predicting labels and scores from embeddings using a pre-trained model.",
# version="1.0.0",
# )
# # Dependency for model access
# def get_model():
# """Dependency to provide the loaded model to endpoints."""
# return tuned_model
# # Define the prediction endpoint
# @app.post("/predict", response_model=PredictionResponse)
# async def predict(
# request: EmbeddingRequest,
# model=Depends(get_model),
# ):
# """
# Predicts the label and score for a given embedding.
# Args:
# request (EmbeddingRequest): A request containing the embedding as a list of floats.
# model: The pre-trained model injected via dependency.
# Returns:
# PredictionResponse: A response containing the predicted label and score.
# """
# try:
# logger.info("Received prediction request.")
# # Convert the input embedding to a DataFrame
# input_data = pd.DataFrame(
# [request.embedding],
# columns=[f"embedding_{i}" for i in range(EMBEDDING_DIMENSION)],
# )
# # Make a prediction using the loaded model
# logger.info("Making prediction...")
# prediction = predict_model(model, data=input_data)
# # Extract the predicted label and score
# predicted_label = prediction["prediction_label"].iloc[0]
# predicted_score = prediction["prediction_score"].iloc[0]
# logger.info(
# f"Prediction successful: label={predicted_label}, score={predicted_score}"
# )
# return PredictionResponse(
# predicted_label=int(predicted_label),
# predicted_score=float(predicted_score),
# )
# except Exception as e:
# logger.error(f"Prediction failed: {str(e)}")
# raise HTTPException(
# status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
# detail=f"An error occurred during prediction: {str(e)}",
# )
# # Health check endpoint
# @app.get("/health", status_code=status.HTTP_200_OK)
# async def health_check():
# """Health check endpoint to verify the API is running."""
# return {"status": "healthy"}
# # Run the FastAPI app
# if __name__ == "__main__":
# import uvicorn
# uvicorn.run(app, host="0.0.0.0", port=8000)
from fastapi import FastAPI, HTTPException, status, Depends
from fastapi.responses import RedirectResponse
from pydantic import BaseModel, conlist, ValidationError
from pydantic_settings import BaseSettings
import pandas as pd
from pycaret.classification import load_model, predict_model
import logging
from typing import Optional, List
import numpy as np
import os
# Configure structured logging
logging.basicConfig(
level=logging.INFO, format="%(asctime)s - %(name)s - %(levelname)s - %(message)s"
)
logger = logging.getLogger(__name__)
# Define settings using Pydantic BaseSettings
class Settings(BaseSettings):
model_path: str = "./api/model/saved_tuned_model"
embedding_dimension: int = 1024
log_level: str = "INFO"
class Config:
env_file = ".env"
env_file_encoding = "utf-8"
settings = Settings()
# Load the saved model
def load_tuned_model(model_path: str):
"""Load the pre-trained model from the specified path."""
try:
logger.info(f"Loading model from {model_path}...")
model = load_model(model_path)
logger.info("Model loaded successfully.")
return model
except Exception as e:
logger.error(f"Failed to load the model: {str(e)}")
raise RuntimeError(f"Model loading failed: {str(e)}")
tuned_model = load_tuned_model(settings.model_path)
# Define the input data model using Pydantic
class EmbeddingRequest(BaseModel):
# embedding: conlist(
# float,
# min_length=settings.embedding_dimension,
# max_length=settings.embedding_dimension,
# )
embedding: List[float]
# Define the response model
class PredictionResponse(BaseModel):
predicted_label: int
predicted_score: float
# Initialize FastAPI app
app = FastAPI(
title="Embedding Prediction API",
description="API for predicting labels and scores from embeddings using a pre-trained model.",
version="1.0.0",
)
# Dependency for model access
def get_model():
"""Dependency to provide the loaded model to endpoints."""
return tuned_model
@app.get("/")
async def root():
return RedirectResponse(url="/docs")
# Define the prediction endpoint
@app.post("/predict", response_model=PredictionResponse)
async def predict(
request: EmbeddingRequest,
model=Depends(get_model),
):
"""
Predicts the label and score for a given embedding.
Args:
request (EmbeddingRequest): A request containing the embedding as a list of floats.
model: The pre-trained model injected via dependency.
Returns:
PredictionResponse: A response containing the predicted label and score.
"""
try:
logger.info("Received prediction request.")
# Convert the input embedding to a DataFrame
input_data = pd.DataFrame(
[request.embedding],
columns=[f"embedding_{i}" for i in range(settings.embedding_dimension)],
)
# Make a prediction using the loaded model
logger.info("Making prediction...")
prediction = predict_model(model, data=input_data)
# Validate the prediction output
if (
"prediction_label" not in prediction.columns
or "prediction_score" not in prediction.columns
):
raise ValueError("Model prediction output is missing required columns.")
# Extract the predicted label and score
predicted_label = prediction["prediction_label"].iloc[0]
predicted_score = prediction["prediction_score"].iloc[0]
if predicted_label == 3:
predicted_label = 4
logger.info(
f"Prediction successful: label={predicted_label}, score={predicted_score}"
)
return PredictionResponse(
predicted_label=int(predicted_label),
predicted_score=float(predicted_score),
)
except ValidationError as e:
logger.error(f"Validation error: {str(e)}")
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"Invalid input data: {str(e)}",
)
except ValueError as e:
logger.error(f"Value error: {str(e)}")
raise HTTPException(
status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
detail=f"Model output validation failed: {str(e)}",
)
except Exception as e:
logger.error(f"Prediction failed: {str(e)}")
raise HTTPException(
status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
detail=f"An error occurred during prediction: {str(e)}",
)
# Health check endpoint
@app.get("/health", status_code=status.HTTP_200_OK)
async def health_check():
"""Health check endpoint to verify the API is running."""
return {"status": "healthy"}
# # Run the FastAPI app
# if __name__ == "__main__":
# import uvicorn
# uvicorn.run(app, host="0.0.0.0", port=8000)
|