به نام خدا
افزایش داده ها (data augmentationata)
افزایش داده ها (data augmentationata)
کدها برگرفته از فصل دو کتاب
و گیت هاب نویسنده کتاب و توسعه دهنده کراس
است.
In [1]:
import os
import matplotlib.pyplot as plt
from keras.preprocessing.image import ImageDataGenerator
In [2]:
# The directory where we will
# store our smaller dataset
base_dir = 'D:/dataset/catDog/catVsdog'
# Directories for our training,
# validation and test splits
train_dir = os.path.join(base_dir, 'train')
validation_dir = os.path.join(base_dir, 'validation')
test_dir = os.path.join(base_dir, 'test')
# Directory with our training cat pictures
train_cats_dir = os.path.join(train_dir, 'cats')
# Directory with our training dog pictures
train_dogs_dir = os.path.join(train_dir, 'dogs')
# Directory with our validation cat pictures
validation_cats_dir = os.path.join(validation_dir, 'cats')
# Directory with our validation dog pictures
validation_dogs_dir = os.path.join(validation_dir, 'dogs')
# Directory with our validation cat pictures
test_cats_dir = os.path.join(test_dir, 'cats')
# Directory with our validation dog pictures
test_dogs_dir = os.path.join(test_dir, 'dogs')
برای بررسی صحت انجام کار تعداد تصاویرآموزشی/آزمون/توسعه را بررسی میکنیم.¶
In [3]:
print('total training cat images:', len(os.listdir(train_cats_dir)))
print('total training dog images:', len(os.listdir(train_dogs_dir)))
print('total validation cat images:', len(os.listdir(validation_cats_dir)))
print('total validation dog images:', len(os.listdir(validation_dogs_dir)))
print('total test cat images:', len(os.listdir(test_cats_dir)))
print('total test dog images:', len(os.listdir(test_dogs_dir)))
In [3]:
datagen = ImageDataGenerator(
rotation_range=40,
width_shift_range=0.2,
height_shift_range=0.2,
shear_range=0.2,
zoom_range=0.2,
horizontal_flip=True,
fill_mode='nearest')
These are just a few of the options available (for more, see the Keras documentation). Let's quickly go over what we just wrote:
rotation_rangeis a value in degrees (0-180), a range within which to randomly rotate pictures.width_shiftandheight_shiftare ranges (as a fraction of total width or height) within which to randomly translate pictures vertically or horizontally.shear_rangeis for randomly applying shearing transformations.zoom_rangeis for randomly zooming inside pictures.horizontal_flipis for randomly flipping half of the images horizontally -- relevant when there are no assumptions of horizontal asymmetry (e.g. real-world pictures).fill_modeis the strategy used for filling in newly created pixels, which can appear after a rotation or a width/height shift.
Let's take a look at our augmented images:
برای درک بهتر نگاهی به تصاویر augment شده می اندازیم:
</div>
</div>
In [10]:
# This is module with image preprocessing utilities
from keras.preprocessing import image
fnames = [os.path.join(train_cats_dir, fname) for fname in os.listdir(train_cats_dir)]
# We pick one image to "augment"
img_path = fnames[9]
# Read the image and resize it
img = image.load_img(img_path, target_size=(150, 150))
# Convert it to a Numpy array with shape (150, 150, 3)
x = image.img_to_array(img)
# Reshape it to (1, 150, 150, 3)
x = x.reshape((1,) + x.shape)
# The .flow() command below generates batches of randomly transformed images.
# It will loop indefinitely, so we need to `break` the loop at some point!
i = 0
for batch in datagen.flow(x, batch_size=1):
plt.figure(i)
imgplot = plt.imshow(image.array_to_img(batch[0]))
i += 1
if i % 4 == 0:
break
plt.show()
If we train a new network using this data augmentation configuration, our network will never see twice the same input. However, the inputs that it sees are still heavily intercorrelated, since they come from a small number of original images -- we cannot produce new information, we can only remix existing information. As such, this might not be quite enough to completely get rid of overfitting.
In [11]:
from keras import layers
from keras import models
from keras import optimizers
model = models.Sequential()
model.add(layers.Conv2D(32, (3, 3), activation='relu',
input_shape=(150, 150, 3)))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(64, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(128, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(128, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Flatten())
model.add(layers.Dropout(0.5))
model.add(layers.Dense(512, activation='relu'))
model.add(layers.Dense(1, activation='sigmoid'))
model.compile(loss='binary_crossentropy',
optimizer=optimizers.RMSprop(lr=1e-4),
metrics=['acc'])
In [10]:
train_datagen = ImageDataGenerator(
rescale=1./255,
rotation_range=40,
width_shift_range=0.2,
height_shift_range=0.2,
shear_range=0.2,
zoom_range=0.2,
horizontal_flip=True,)
# Note that the validation data should not be augmented!
test_datagen = ImageDataGenerator(rescale=1./255)
train_generator = train_datagen.flow_from_directory(
# This is the target directory
train_dir,
# All images will be resized to 150x150
target_size=(150, 150),
batch_size=32,
# Since we use binary_crossentropy loss, we need binary labels
class_mode='binary')
validation_generator = test_datagen.flow_from_directory(
validation_dir,
target_size=(150, 150),
batch_size=32,
class_mode='binary')
history = model.fit_generator(
train_generator,
steps_per_epoch=100,
epochs=100,
validation_data=validation_generator,
validation_steps=50)
In [11]:
model.save('cats_and_dogs_small_2.h5')
رسم نمودار
</div>
</div>
In [12]:
acc = history.history['acc']
val_acc = history.history['val_acc']
loss = history.history['loss']
val_loss = history.history['val_loss']
epochs = range(len(acc))
plt.plot(epochs, acc, 'bo', label='Training acc')
plt.plot(epochs, val_acc, 'b', label='Validation acc')
plt.title('Training and validation accuracy')
plt.legend()
plt.figure()
plt.plot(epochs, loss, 'bo', label='Training loss')
plt.plot(epochs, val_loss, 'b', label='Validation loss')
plt.title('Training and validation loss')
plt.legend()
plt.show()
دوره مقدماتی یادگیری عمیق
علیرضا اخوان پور
پنج شنبه، ۲۵ بهمن ۱۳۹۷
Class.Vision - AkhavanPour.ir - GitHub
علیرضا اخوان پور
پنج شنبه، ۲۵ بهمن ۱۳۹۷
In [ ]: