|
| 1 | +{pip install flask, |
| 2 | + "image": "mcr.microsoft.com/devcontainers/universal:2", |
| 3 | + "features": {}from flask import Flask, request, jsonify |
| 4 | +import pandas as pd |
| 5 | +from sklearn.preprocessing import StandardScaler |
| 6 | + |
| 7 | +# Initialize Flask app |
| 8 | +app = Flask(__name__) |
| 9 | + |
| 10 | +# Pre-trained model (make sure to load your actual trained model here) |
| 11 | +from sklearn.ensemble import RandomForestClassifier |
| 12 | + |
| 13 | +# Load model (replace this with your actual model loading process) |
| 14 | +# model = load_your_trained_model() |
| 15 | + |
| 16 | +# For example, if using a Random Forest: |
| 17 | +model = RandomForestClassifier(n_estimators=100, random_state=42) |
| 18 | + |
| 19 | +# Pre-process input data (standardize it based on your training process) |
| 20 | +scaler = StandardScaler() |
| 21 | + |
| 22 | +@app.route('/predict', methods=['POST']) |
| 23 | +def predict(): |
| 24 | + # Get data from the request |
| 25 | + data = request.get_json(force=True) |
| 26 | + |
| 27 | + # Ensure the data has the correct format (this will depend on your feature set) |
| 28 | + features = pd.DataFrame(data['features']) |
| 29 | + |
| 30 | + # Standardize features |
| 31 | + features = scaler.transform(features) |
| 32 | + |
| 33 | + # Make prediction |
| 34 | + prediction = model.predict(features) |
| 35 | + |
| 36 | + # Return the prediction as JSON |
| 37 | + return jsonify({'prediction': prediction.tolist()}) |
| 38 | + |
| 39 | +if __name__ == '__main__': |
| 40 | + app.run(debug=True) |
| 41 | +}curl -X POST http://127.0.0.1:5000/predict -H "Content-Type: application/json" -d '{"features": [[1.5, 2.3, 3.1]]}' |
0 commit comments