lunny/keras
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Finish PR backporting

Browse Source
This commit is contained in:
Francois Chollet 2016-04-04 11:30:24 -07:00
parent eb8b40cccd
commit 2902149f77
6 changed files with 19 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
examples/antirectifier.py
View File

@ -63,7 +63,7 @@ batch_size = 128
nb_classes = 10 nb_classes = 10
nb_epoch = 40 nb_epoch = 40
# the data, shuffled and split between tran and test sets # the data, shuffled and split between train and test sets
(X_train, y_train), (X_test, y_test) = mnist.load_data() (X_train, y_train), (X_test, y_test) = mnist.load_data()
X_train = X_train.reshape(60000, 784) X_train = X_train.reshape(60000, 784)

2
examples/mnist_cnn.py
View File

@ -28,7 +28,7 @@ nb_pool = 2
# convolution kernel size # convolution kernel size
nb_conv = 3 nb_conv = 3
# the data, shuffled and split between tran and test sets # the data, shuffled and split between train and test sets
(X_train, y_train), (X_test, y_test) = mnist.load_data() (X_train, y_train), (X_test, y_test) = mnist.load_data()
X_train = X_train.reshape(X_train.shape[0], 1, img_rows, img_cols) X_train = X_train.reshape(X_train.shape[0], 1, img_rows, img_cols)

2
examples/mnist_mlp.py
View File

@ -20,7 +20,7 @@ batch_size = 128
nb_classes = 10 nb_classes = 10
nb_epoch = 20 nb_epoch = 20
# the data, shuffled and split between tran and test sets # the data, shuffled and split between train and test sets
(X_train, y_train), (X_test, y_test) = mnist.load_data() (X_train, y_train), (X_test, y_test) = mnist.load_data()
X_train = X_train.reshape(60000, 784) X_train = X_train.reshape(60000, 784)

2
examples/mnist_siamese_graph.py
View File

@ -73,7 +73,7 @@ def compute_accuracy(predictions, labels):
return labels[predictions.ravel() < 0.5].mean() return labels[predictions.ravel() < 0.5].mean()
# the data, shuffled and split between tran and test sets # the data, shuffled and split between train and test sets
(X_train, y_train), (X_test, y_test) = mnist.load_data() (X_train, y_train), (X_test, y_test) = mnist.load_data()
X_train = X_train.reshape(60000, 784) X_train = X_train.reshape(60000, 784)
X_test = X_test.reshape(10000, 784) X_test = X_test.reshape(10000, 784)

7
keras/engine/training.py
View File

@ -3,6 +3,7 @@
from __future__ import print_function from __future__ import print_function
from __future__ import absolute_import from __future__ import absolute_import
import warnings
import copy import copy
import time import time
import numpy as np import numpy as np
@ -1170,6 +1171,12 @@ class Model(Container):
samples_seen += batch_size samples_seen += batch_size
# epoch finished # epoch finished
if samples_seen > samples_per_epoch:
warnings.warn('Epoch comprised more than '
'`samples_per_epoch` samples, '
'which might affect learning results. '
'Set `samples_per_epoch` correctly '
'to avoid this warning.')
if samples_seen >= samples_per_epoch and do_validation: if samples_seen >= samples_per_epoch and do_validation:
if val_gen: if val_gen:
val_outs = self.evaluate_generator(validation_data, val_outs = self.evaluate_generator(validation_data,

17
keras/preprocessing/image.py
View File

@ -11,14 +11,13 @@ from scipy import linalg
from os import listdir from os import listdir
from os.path import isfile, join from os.path import isfile, join
import random
import math import math
from six.moves import range from six.moves import range
import threading import threading
def random_rotation(x, rg, fill_mode='nearest', cval=0.): def random_rotation(x, rg, fill_mode='nearest', cval=0.):
angle = random.uniform(-rg, rg) angle = np.random.uniform(-rg, rg)
x = ndimage.interpolation.rotate(x, angle, x = ndimage.interpolation.rotate(x, angle,
axes=(1, 2), axes=(1, 2),
reshape=False, reshape=False,
@ -31,9 +30,9 @@ def random_shift(x, wrg, hrg, fill_mode='nearest', cval=0.):
shift_x = shift_y = 0 shift_x = shift_y = 0
if wrg: if wrg:
shift_x = random.uniform(-wrg, wrg) * x.shape[2] shift_x = np.random.uniform(-wrg, wrg) * x.shape[2]
if hrg: if hrg:
shift_y = random.uniform(-hrg, hrg) * x.shape[1] shift_y = np.random.uniform(-hrg, hrg) * x.shape[1]
x = ndimage.interpolation.shift(x, (0, shift_y, shift_x), x = ndimage.interpolation.shift(x, (0, shift_y, shift_x),
order=0, order=0,
mode=fill_mode, mode=fill_mode,
@ -59,7 +58,7 @@ def random_barrel_transform(x, intensity):
def random_shear(x, intensity, fill_mode='nearest', cval=0.): def random_shear(x, intensity, fill_mode='nearest', cval=0.):
shear = random.uniform(-intensity, intensity) shear = np.random.uniform(-intensity, intensity)
shear_matrix = np.array([[1.0, -math.sin(shear), 0.0], shear_matrix = np.array([[1.0, -math.sin(shear), 0.0],
[0.0, math.cos(shear), 0.0], [0.0, math.cos(shear), 0.0],
[0.0, 0.0, 1.0]]) [0.0, 0.0, 1.0]])
@ -76,8 +75,8 @@ def random_channel_shift(x, rg):
def random_zoom(x, rg, fill_mode='nearest', cval=0.): def random_zoom(x, rg, fill_mode='nearest', cval=0.):
zoom_w = random.uniform(1.-rg, 1.) zoom_w = np.random.uniform(1.-rg, 1.)
zoom_h = random.uniform(1.-rg, 1.) zoom_h = np.random.uniform(1.-rg, 1.)
x = ndimage.interpolation.zoom(x, zoom=(1., zoom_w, zoom_h), x = ndimage.interpolation.zoom(x, zoom=(1., zoom_w, zoom_h),
mode=fill_mode, mode=fill_mode,
cval=cval) cval=cval)
@ -253,10 +252,10 @@ class ImageDataGenerator(object):
if self.width_shift_range or self.height_shift_range: if self.width_shift_range or self.height_shift_range:
x = random_shift(x, self.width_shift_range, self.height_shift_range) x = random_shift(x, self.width_shift_range, self.height_shift_range)
if self.horizontal_flip: if self.horizontal_flip:
if random.random() < 0.5: if np.random.random() < 0.5:
x = horizontal_flip(x) x = horizontal_flip(x)
if self.vertical_flip: if self.vertical_flip:
if random.random() < 0.5: if np.random.random() < 0.5:
x = vertical_flip(x) x = vertical_flip(x)
if self.shear_range: if self.shear_range:
x = random_shear(x, self.shear_range) x = random_shear(x, self.shear_range)