import tensorflow as tf
from tensorflow.keras.preprocessing.image import ImageDataGenerator

Set the seed

tf.random.set_seed(42)

Preprocess data(get all of the pixel values between 0 & 1, also calles scaling / normalization)

train_datagen = ImageDataGenerator(rescale = 1./255)
valid_datagen = ImageDataGenerator(rescale = 1./255) 

Setup path to our data directories

train_dir = '/content/pizza_steak/train'
test_dir = 'pizza_steak/test'

Import data from directories and turn it into batches

train_data = train_datagen.flow_from_directory(directory = train_dir,
                                               batch_size = 32,
                                               target_size = (224, 224),
                                               class_mode = 'binary',
                                               seed = 42)

valid_data = valid_datagen.flow_from_directory(directory = test_dir,
                                               batch_size = 32,
                                               target_size = (224, 224),
                                               class_mode = 'binary',
                                               seed = 42)

Build a CNN model (same as the Tiny VGG on the CNN explain website)

model_1 = tf.keras.models.Sequential([
tf.keras.layers.Conv2D(filters = 10,
                       kernel_size = 3,
                       activation = 'relu',
                       input_shape = (244, 244, 3)),

tf.keras.layers.Conv2D(10, 3, activation = 'relu'),
tf.keras.layers.MaxPool2D(pool_size = 2,
                          padding = 'valid'),

tf.keras.layers.Conv2D(10, 3, activation = 'relu'),
tf.keras.layers.Conv2D(10, 3, activation = 'relu'),
tf.keras.layers.MaxPool2D(2),

tf.keras.layers.Flatten(),

tf.keras.layers.Dense(1, activation = 'sigmoid')
])

Compile our CNN

model_1.compile(
loss = 'binary_crossentropy',
optimizer = tf.keras.optimizers.Adam(),
metrics = ['accuracy']
)

Fit the model

history_1 = model_1.fit(train_data,
                        epochs = 5,
                        steps_per_epoch = len(train_data),
                        validation_data = valid_data,
                        validation_steps = len(valid_data))

CodePudding user response：

You either need to change your input_shape to (224, 224, 3) or the target_size to (244, 244, 3). It will not work with different shapes. Here is a working example:

import tensorflow as tf
import matplotlib.pyplot as plt

BATCH_SIZE = 32

flowers = tf.keras.utils.get_file(
    'flower_photos',
    'https://storage.googleapis.com/download.tensorflow.org/example_images/flower_photos.tgz',
    untar=True)

train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(rescale=1./255)


train_data = train_datagen.flow_from_directory(directory = flowers,
                                               batch_size = 32,
                                               target_size = (224, 224),
                                               class_mode = 'sparse',
                                               seed = 42)


model_1 = tf.keras.models.Sequential([
    tf.keras.layers.Conv2D(filters = 10,
                          kernel_size = 3,
                          activation = 'relu',
                          input_shape = (224, 224, 3)),
    tf.keras.layers.Conv2D(10, 3, activation = 'relu'),
    tf.keras.layers.MaxPool2D(pool_size = 2,
                              padding = 'valid'),

    tf.keras.layers.Conv2D(10, 3, activation = 'relu'),
    tf.keras.layers.Conv2D(10, 3, activation = 'relu'),
    tf.keras.layers.MaxPool2D(2),
    tf.keras.layers.Flatten(),
    tf.keras.layers.Dense(5, activation = 'softmax')
])

model_1.compile(
loss = 'sparse_categorical_crossentropy',
optimizer = tf.keras.optimizers.Adam(),
metrics = ['accuracy']
)
history_1 = model_1.fit(train_data,steps_per_epoch = len(train_data),
                        epochs = 5)