--- library_name: keras tags: - Image Classification --- # Cifar CNN with Adversarial Training (Teeny-Tiny Castle) This model is part of a tutorial tied to the [Teeny-Tiny Castle](https://github.com/Nkluge-correa/TeenyTinyCastle), an open-source repository containing educational tools for AI Ethics and Safety research. ## How to Use ```python !pip install huggingface_hub["tensorflow"] -q import numpy as np import tensorflow as tf import matplotlib.pyplot as plt from huggingface_hub import from_pretrained_keras # Download the CIFAR-10 dataset (x_train, y_train), (x_test, y_test) = tf.keras.datasets.cifar10.load_data() class_names = ['Airplane', 'Automobile', 'Bird', 'Cat', 'Deer', 'Dog', 'Frog', 'Horse', 'Ship', 'Truck'] plt.figure(figsize=[10, 10]) for i in range(25): plt.subplot(5, 5, i+1) plt.xticks([]) plt.yticks([]) plt.grid(False) plt.imshow(x_test[i], cmap=plt.cm.binary) plt.xlabel(class_names[y_test[i][0]]) plt.show() # Load the model from the Hub model = from_pretrained_keras("AiresPucrs/Cifar-CNN-with-adversarial-training") model.compile( loss=tf.keras.losses.CategoricalCrossentropy(), metrics=['categorical_accuracy'] ) x_train = x_train.astype('float32') x_train = x_train / 255. y_train = tf.keras.utils.to_categorical(y_train, 10) x_test = x_test.astype('float32') x_test = x_test / 255. y_test = tf.keras.utils.to_categorical(y_test, 10) test_loss_score, test_acc_score = model.evaluate(x_test, y_test, verbose=0) model.summary() print(f'Loss: {round(test_loss_score, 2)}.') print(f'Accuracy: {round(test_acc_score * 100, 2)} %.') ```