# Image Classifier with TensorFlow

This guide demonstrates how to build and train a Convolutional Neural Network (CNN) to classify images using TensorFlow and Keras. We will use the CIFAR-10 dataset, which consists of 60,000 32x32 color images in 10 classes.

**Modules Used:**
*   `tensorflow`: An open-source platform for machine learning.
*   `numpy`: For numerical operations.
*   [[programming/python/modules/matplotlib-module|matplotlib]]: To visualize the data and training results.

## Installation

```bash
pip install tensorflow numpy matplotlib
```

## The Code

Save this as `image_classifier.py`.

```python
import tensorflow as tf
from tensorflow.keras import datasets, layers, models
import matplotlib.pyplot as plt
import numpy as np

def run_classifier():
    # 1. Load and Preprocess Data
    print("Loading CIFAR-10 dataset...")
    (train_images, train_labels), (test_images, test_labels) = datasets.cifar10.load_data()

    # Normalize pixel values to be between 0 and 1
    train_images, test_images = train_images / 255.0, test_images / 255.0

    class_names = ['airplane', 'automobile', 'bird', 'cat', 'deer',
                   'dog', 'frog', 'horse', 'ship', 'truck']

    # 2. Build the CNN Model
    print("Building model...")
    model = models.Sequential()
    # Convolutional layers to extract features
    model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(32, 32, 3)))
    model.add(layers.MaxPooling2D((2, 2)))
    model.add(layers.Conv2D(64, (3, 3), activation='relu'))
    model.add(layers.MaxPooling2D((2, 2)))
    model.add(layers.Conv2D(64, (3, 3), activation='relu'))
    
    # Dense layers for classification
    model.add(layers.Flatten())
    model.add(layers.Dense(64, activation='relu'))
    model.add(layers.Dense(10)) # 10 output classes

    model.summary()

    # 3. Compile the Model
    model.compile(optimizer='adam',
                  loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
                  metrics=['accuracy'])

    # 4. Train the Model
    print("Training model (this may take a while)...")
    history = model.fit(train_images, train_labels, epochs=10, 
                        validation_data=(test_images, test_labels))

    # 5. Evaluate
    print("\nEvaluating model...")
    test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)
    print(f"\nTest accuracy: {test_acc:.4f}")

    # 6. Plot Training History
    plt.plot(history.history['accuracy'], label='accuracy')
    plt.plot(history.history['val_accuracy'], label = 'val_accuracy')
    plt.xlabel('Epoch')
    plt.ylabel('Accuracy')
    plt.ylim([0.5, 1])
    plt.legend(loc='lower right')
    plt.title("Training History")
    plt.show()

if __name__ == "__main__":
    run_classifier()
```

## Usage

```bash
python image_classifier.py
```

Note: Training a CNN on a CPU can be slow. If you have a compatible NVIDIA GPU, ensure you have installed the necessary CUDA drivers and the `tensorflow[and-cuda]` package for faster training.

[[programming/python/python]]