| import joblib | |
| import numpy as np | |
| # Load model and label encoder | |
| model = joblib.load("soil.pkl") | |
| label_encoder = joblib.load("label_encoder.pkl") | |
| def predict(inputs): | |
| """ | |
| Inputs: List of 7 features [N, P, K, temperature, humidity, ph, rainfall] | |
| """ | |
| input_array = np.array(inputs).reshape(1, -1) | |
| prediction = model.predict(input_array) | |
| crop = label_encoder.inverse_transform(prediction) | |
| return crop[0] | |