import os
import warnings
warnings.filterwarnings("ignore")
import tensorflow as tf
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.applications.resnet import ResNet50
from pathlib import Path
import numpy as np#数据所在文件夹
base_dir = './data/cats_and_dogs'
train_dir = Path(os.path.join(base_dir,'train'))
file_pattern = os.path.join(train_dir,'*/*.jpg')
image_count = len(list(train_dir.glob('*/*.jpg')))list_ds = tf.data.Dataset.list_files(file_pattern,shuffle = False)
list_ds = list_ds.shuffle(image_count, reshuffle_each_iteration=False)
for f in list_ds.take(5):print(f.numpy())class_names = np.array(sorted([item.name for item in train_dir.glob('*') ]))
print(class_names)val_size = int(image_count * 0.2)
train_data = list_ds.skip(val_size)
validation_data = list_ds.take(val_size)
print(tf.data.experimental.cardinality(train_data).numpy())
print(tf.data.experimental.cardinality(validation_data).numpy())def get_label(file_path):parts = tf.strings.split(file_path, os.path.sep)one_hot = parts[-2] == class_namesreturn tf.argmax(one_hot)def decode_img(img):img = tf.io.decode_jpeg(img, channels=3)return tf.image.resize(img, [64, 64])def process_path(file_path):label = get_label(file_path)img = tf.io.read_file(file_path)img = decode_img(img)return img, labeltrain_data = train_data.map(process_path, num_parallel_calls=tf.data.AUTOTUNE)
validation_data = validation_data.map(process_path, num_parallel_calls=tf.data.AUTOTUNE)for image, label in train_data.take(2):print("Image shape: ", image.numpy().shape)print("Label: ", label.numpy())def configure_for_performance(ds):ds = ds.cache()ds = ds.shuffle(buffer_size=1000)ds = ds.batch(4)ds = ds.prefetch(buffer_size=tf.data.AUTOTUNE)return dstrain_data = configure_for_performance(train_data)
validation_data = configure_for_performance(validation_data)save_model_cb = tf.keras.callbacks.ModelCheckpoint(filepath='model_resnet50_cats_and_dogs.h5', save_freq='epoch')base_model = ResNet50(weights='imagenet', include_top=False, input_shape=(64, 64, 3))
base_model.trainable = Truemodel = tf.keras.models.Sequential([base_model,tf.keras.layers.Dropout(0.2),tf.keras.layers.Flatten(),tf.keras.layers.Dense(512, activation='relu',kernel_regularizer=tf.keras.regularizers.l2(l=0.01)),tf.keras.layers.Dense(1, activation='sigmoid')
])model.compile(loss='binary_crossentropy',optimizer = Adam(lr=1e-3),metrics = ['acc'])history = model.fit(train_data.repeat(),steps_per_epoch=100,epochs=50,validation_data=validation_data.repeat(),validation_steps=50,verbose=1,callbacks = [save_model_cb])