#!/bin/python3
# A small example of using the Keras library to create a neoral network
# and train it on a data set created randomly using NumPy.
# AuthoR: Dana Vrajitoru and ...
# Class: C462/B551/I400 Artificial Intelligence
# Updated: Fall 2025

import keras
import os
from keras.models import Sequential
from keras.layers import Dense
import numpy as np

#print(keras.__version__)
os.environ["KERAS_BACKEND"] = "tensorflow"

def test_sequential():
    # 1. Generate synthetic data
    # 1000 samples, 2 features
    X = np.random.rand(1000, 2) * 10
    print("Input: ", X)

    # Create labels based on a simple rule: if sum of features > 10,
    # class 1, else class 0
    y = (X[:, 0] + X[:, 1] > 10).astype(int)
    print("Output: ", y)

    # 2. Define the Keras model using the Sequential API
    model = Sequential()
    # Add a hidden layer with 10 neurons and ReLU activation
    model.add(Dense(10, input_dim=2, activation='relu'))
    # Add the output layer with 1 neuron (for binary classification)
    # and sigmoid activation
    model.add(Dense(1, activation='sigmoid'))
    
    # 3. Compile the model
    # Use binary cross-entropy for binary classification, Adam
    # optimizer, and accuracy as a metric
    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])

    # 4. Train the model
    # Train for 50 epochs with a batch size of 32
    model.fit(X, y, epochs=50, batch_size=32, verbose=0) # verbose=0 to suppress training output

    # 5. Evaluate the model
    loss, accuracy = model.evaluate(X, y, verbose=0)
    print(f"Model Accuracy: {accuracy*100:.2f}%")

    # 6. Make predictions
    # Predict on a few new samples
    new_samples = np.array([[1, 2], [8, 9], [3, 7], [1, 9]])
    predictions = model.predict(new_samples)
    print("\nPredictions for new samples:")
    for i, sample in enumerate(new_samples):
        predicted_class = int(predictions[i][0] > 0.5) # Convert probability to binary class
        print(f"Sample: {sample}, Predicted probability: {predictions[i][0]:.2f}, Predicted class: {predicted_class}")

if __name__ == '__main__':
    test_sequential()