CNN

import os

os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0'

import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
from tensorflow import keras
from tensorflow.keras import layers



from keras.src.utils import image_dataset_from_directory

train_dataset = image_dataset_from_directory(
    "C:/Users/laure/code/pycharmProjects/BirdDataset/train",
    image_size=(224, 224),
    batch_size=32)
test_dataset = image_dataset_from_directory(
    "C:/Users/laure/code/pycharmProjects/BirdDataset/test",
    image_size=(224, 224),
    batch_size=32)
validata = image_dataset_from_directory(
    "C:/Users/laure/code/pycharmProjects/BirdDataset/valid",
    image_size=(224, 224),
    batch_size=32)

class_names = train_dataset.class_names
print(class_names)

data_augmentation = keras.Sequential(
    [
        layers.RandomFlip("horizontal"),
        layers.RandomRotation(0.2),
        layers.RandomZoom(0.4),
    ]
)

img_size = (224, 224)

inputs = keras.Input(shape=(224, 224, 3))
x = data_augmentation(inputs)
x = layers.Rescaling(1./255)(inputs)

x = layers.BatchNormalization()(x)
x = layers.Conv2D(filters=32, kernel_size=3, activation="relu")(x)
x = layers.MaxPooling2D(pool_size=2, padding="same")(x)

x = layers.BatchNormalization()(x)
x = layers.Conv2D(filters=64, kernel_size=3, activation="relu")(x)
x = layers.MaxPooling2D(pool_size=2, padding="same")(x)

x = layers.BatchNormalization()(x)
x = layers.Conv2D(filters=128, kernel_size=3, activation="relu")(x)
x = layers.MaxPooling2D(pool_size=2, padding="same")(x)

x = layers.BatchNormalization()(x)
x = layers.Conv2D(filters=256, kernel_size=3, activation="relu")(x)
x = layers.MaxPooling2D(pool_size=2, padding="same")(x)

x = layers.BatchNormalization()(x)
x = layers.Conv2D(filters=256, kernel_size=3, activation="relu")(x)
x = layers.GlobalAveragePooling2D()(x)

x = layers.Dropout(0.5)(x)
outputs = layers.Dense(525, activation="softmax")(x)
model = keras.Model(inputs=inputs, outputs=outputs)
model.summary()

model.compile(loss="sparse_categorical_crossentropy",
              optimizer="adam",
              metrics=["accuracy"])

callbacks = [
    keras.callbacks.ModelCheckpoint(
        filepath="bird525.keras",
        save_best_only=True,
        monitor="val_loss")]

history = model.fit(
    train_dataset,
    epochs=20,
    validation_data=validata,
    callbacks=callbacks)