python 24 lines · 1 tab

Serving scikit-learn models behind a FastAPI prediction API

Dr. Elena Vasquez Apr 2026

1 tab

import joblib
import pandas as pd
from fastapi import FastAPI
from pydantic import BaseModel

app = FastAPI(title='Churn Prediction API')
model = joblib.load('artifacts/churn_pipeline.joblib')

class PredictionRequest(BaseModel):
    age: int
    income: float
    country: str
    plan_tier: str
    days_since_last_login: int

@app.get('/health')
def health():
    return {'status': 'ok'}

@app.post('/predict')
def predict(payload: PredictionRequest):
    frame = pd.DataFrame([payload.model_dump()])
    probability = float(model.predict_proba(frame)[:, 1][0])
    return {'churn_probability': probability, 'model_version': '2026-04-07'}

1 file · python Explain with highlit

Deployment should not rewrite the feature logic from scratch. I expose trained pipelines behind FastAPI so the exact preprocessing and estimator objects travel together. Strong request schemas and explicit model versioning keep this boring in the right way.

Related snips

python

import mlflow
import mlflow.sklearn
from sklearn.metrics import roc_auc_score

mlflow.set_experiment('customer-churn')

Experiment tracking and model registry workflows with MLflow

mlflow experiment-tracking model-registry

by Dr. Elena Vasquez 1 tab

python

from sklearn.linear_model import LinearRegression, Ridge, Lasso, ElasticNet
from sklearn.metrics import mean_absolute_error, root_mean_squared_error

models = {
    'linear': LinearRegression(),
    'ridge': Ridge(alpha=1.0),

Regression workflows with linear, ridge, lasso, and elastic net

scikit-learn regression ridge

by Dr. Elena Vasquez 1 tab

python

import pandas as pd
from sklearn.compose import ColumnTransformer
from sklearn.impute import SimpleImputer
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import OneHotEncoder, StandardScaler

Encoding categorical variables without creating leakage

categorical-encoding preprocessing scikit-learn

by Dr. Elena Vasquez 1 tab

python

from sklearn.model_selection import GridSearchCV, RandomizedSearchCV
from sklearn.ensemble import RandomForestClassifier

grid_search = GridSearchCV(
    estimator=RandomForestClassifier(random_state=42, n_jobs=-1),
    param_grid={

Hyperparameter tuning with GridSearchCV and randomized search

hyperparameter-tuning gridsearch scikit-learn

by Dr. Elena Vasquez 1 tab

python

from typing import Any, Optional

from fastapi.responses import JSONResponse
from pydantic import BaseModel

Consistent JSON Error Responses in FastAPI With a Custom Exception Handler

fastapi error-handling exceptions

by codesnips 3 tabs

python

from typing import Optional
from pydantic import BaseModel, EmailStr, Field, field_validator, model_validator


class CreateUserRequest(BaseModel):
    model_config = {"extra": "forbid"}

Validating JSON Payloads with Pydantic v2 and Returning Field-Level Errors in FastAPI

fastapi pydantic validation

by codesnips 3 tabs

Share this code

Here's the card — post it anywhere.

Serving scikit-learn models behind a FastAPI prediction API — share card

Link copied