Machine Learning Model Deployment with FastAPI

Why FastAPI for ML Deployment?

FastAPI is a modern, fast web framework for building APIs with Python 3.7+ based on standard Python type hints. It's particularly well-suited for deploying machine learning models due to its performance and ease of use.

Key Features

• Fast: Very high performance, on par with NodeJS and Go
• Fast to code: Increase development speed by 200-300%
• Fewer bugs: Reduce human-induced errors by 40%
• Intuitive: Great editor support with auto-completion
• Short: Minimize code duplication

Basic FastAPI Application

from fastapi import FastAPI

app = FastAPI()

@app.get("/")
async def read_root():
    return {"Hello": "World"}

@app.get("/items/{item_id}")
async def read_item(item_id: int, q: str = None):
    return {"item_id": item_id, "q": q}

Loading and Using ML Models

FastAPI can easily serve machine learning models. Here's an example using scikit-learn and joblib.

from fastapi import FastAPI
from pydantic import BaseModel
import joblib
import numpy as np

app = FastAPI()

# Load the trained model
model = joblib.load('model.pkl')

class PredictionRequest(BaseModel):
    features: list

@app.post("/predict")
async def predict(request: PredictionRequest):
    # Make prediction
    prediction = model.predict([request.features])
    return {"prediction": prediction.tolist()}

Model Serialization

Before deploying, you need to serialize your trained model. Popular options include:

• joblib (for scikit-learn models)
• pickle (Python standard library)
• ONNX (framework-agnostic format)

import joblib

# Save model
joblib.dump(model, 'model.pkl')

# Load model
loaded_model = joblib.load('model.pkl')

FastAPI provides an excellent platform for deploying machine learning models with high performance and developer-friendly features.