train2.py

import cv2
import os
import numpy as np
from huggingface_hub import push_to_hub_keras
from tensorflow.keras.applications import VGG16
from tensorflow.keras.layers import Flatten, Dense, Dropout
from tensorflow.keras.models import Model
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras.applications.vgg16 import preprocess_input  # VGG16 specific pre-processing
from PIL import Image  # For grayscale to RGB conversion
# Early Stopping to prevent overfitting
from tensorflow.keras.callbacks import EarlyStopping
# Compile the model with a learning rate scheduler for potential improvement
from tensorflow.keras.callbacks import LearningRateScheduler

sac = os.getenv('accesstoken')
# Define data paths and hyperparameters
train_data_dir = 'datasets'
validation_data_dir = 'first_ten_files'
test_data_dir = 'first_ten_files'
img_size = (224, 224)  # Adjust based on pre-trained model input size
img_height, img_width = 224, 224
batch_size = 32
num_classes = 10  # Adjust based on your fingerprint classification types
# Data augmentation (optional)
train_datagen = ImageDataGenerator(rescale=1./255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True)
val_datagen = ImageDataGenerator(rescale=1./255)

# Load training and validation data
train_generator = train_datagen.flow_from_directory(
    train_data_dir,
    target_size=(img_height, img_width),
    batch_size=32,
    class_mode='categorical'
)
val_generator = val_datagen.flow_from_directory(
    validation_data_dir,
    target_size=(img_height, img_width),
    batch_size=32,
    class_mode='categorical'
)

# Load pre-trained model (VGG16 in this example)
base_model = VGG16(weights="imagenet", include_top=False, input_shape=(img_height, img_width, 3))
base_model.trainable = False  # Freeze pre-trained layers

# Add new layers
x = base_model.output
x = Flatten()(x)
x = Dense(1024, activation='relu')(x)  # Add a dense layer with 1024 neurons and ReLU activation
x = Dropout(0.5)(x)  # Add dropout for regularization
predictions = Dense(10, activation='softmax')(x)  # Output layer with 10 units for 10 classes

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

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

# Early stopping callback
early_stopping = EarlyStopping(monitor='val_loss', patience=3)

# Train the model
model.fit(
    train_generator,
    epochs=10,
    validation_data=val_generator,
    callbacks=[early_stopping]
)

# # Evaluate the model on unseen data
# test_generator = val_datagen.flow_from_directory(
#     test_data_dir,
#     target_size=img_size,
#     batch_size=batch_size,
#     class_mode='categorical',
#     color_mode='grayscale'
# )

# test_loss, test_acc = model.evaluate(test_generator)
# print('Test accuracy:', test_acc)

# Save the model for future use
model.save('fingerprint_classifier.keras')
labels = train_generator.class_indices
labels = dict((v, k) for k, v in labels.items())
print(labels)
# Save the trained model
# model.save('fingerprint_recognition_model.keras')
# Show the model architecture
model.summary()
push_to_hub_keras(
    model,
    repo_id="okeowo1014/fingerprintgrecognizer",
    commit_message="Initial commit model",
    token=sac,
    include_optimizer=True,
)
print(f"Model pushed to Hugging Face Hub")