image-attribute.py

import tensorflow as tf
from tensorflow.keras.applications import ResNet50
from tensorflow.keras.layers import Dense, GlobalAveragePooling2D
from tensorflow.keras.models import Model
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.preprocessing.image import ImageDataGenerator

# Set the paths to your DeepFashion dataset
train_data_dir = 'data/train'
validation_data_dir = 'data/validation'

#
# Set the number of classes and batch size
num_classes = 50
batch_size = 32
#
# Data augmentation and normalization
train_datagen = ImageDataGenerator(
    rescale=1.0 / 255,
    shear_range=0.2,
    zoom_range=0.2,
    horizontal_flip=True
)

validation_datagen = ImageDataGenerator(rescale=1.0 / 255)
#
# Load the ResNet50 model without the top layer
base_model = ResNet50(weights='imagenet', include_top=False, input_shape=(224, 224, 3))

# Add a global average pooling layer
x = base_model.output
x = GlobalAveragePooling2D()(x)

# Add a fully connected layer with 1024 units
x = Dense(1024, activation='relu')(x)

# Add the final output layer with the number of classes
predictions = Dense(num_classes, activation='sigmoid')(x)

# Create the model
model = Model(inputs=base_model.input, outputs=predictions)

# Freeze the base model layers
for layer in base_model.layers:
    layer.trainable = False

# Compile the model
model.compile(optimizer=Adam(lr=0.001), loss='binary_crossentropy', metrics=['accuracy'])

# Load and preprocess the training and validation data
train_generator = train_datagen.flow_from_directory(
    train_data_dir,
    target_size=(224, 224),
    batch_size=batch_size,
    class_mode='categorical'
)

validation_generator = validation_datagen.flow_from_directory(
    validation_data_dir,
    target_size=(224, 224),
    batch_size=batch_size,
    class_mode='categorical'
)

# Train the model
model.fit(
    train_generator,
    steps_per_epoch=train_generator.samples,
    epochs=10,
    validation_data=validation_generator,
    validation_steps=validation_generator.samples
)

# Save the trained model
model.save('deepfashion_attribute_model.h5')