!git clone https://github.com/cbtn-data-science-ml/introduction-to-deep-learning.git

import os

%pwd

os.chdir('introduction-to-deep-learning')

%pwd

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

# Reproduceable random
tf.random.set_seed(42)

# Data Gens
train_datagen = ImageDataGenerator(rescale=1./255)
test_datagen = ImageDataGenerator(rescale=1./255)

# Train data generators
train_generator = train_datagen.flow_from_directory(
    'ramen_sushi/train/', # path to train data
    target_size=(200,200),
    batch_size=32,
    class_mode='binary',
    seed=42
)

test_generator = test_datagen.flow_from_directory(
    'ramen_sushi/test/', # path to train data
    target_size=(200,200),
    batch_size=32,
    class_mode='binary',
    seed=42
)

# Define model
model = tf.keras.Sequential([
    tf.keras.layers.Flatten(input_shape=(200,200, 3)),
    tf.keras.layers.Dense(32, activation='relu'),
    tf.keras.layers.Dense(1, activation='sigmoid')
])

# Compile model
model.compile(optimizer='adam',
               loss='binary_crossentropy',
               metrics=['accuracy'])

# Train model
history = model.fit(train_generator, epochs=5)

model.summary()