def test_getRandomBatch():
"""Batch tensors should have the correct shape"""
batchSize = 22
X, y = qrcodes.getRandomBatch(batchSize)
y = np.asarray(y)
assert_equals((batchSize, qrcodes.IMAGE_SIZE, qrcodes.IMAGE_SIZE, 1),
X.shape)
assert_equals((batchSize, len(qrcodes.CHARACTER_SET)), y.shape)
def weight_variable(shape):
initial = tf.truncated_normal(shape, stddev=0.1, name="weights")
return tf.Variable(initial)
def bias_variable(shape):
initial = tf.constant(0.1, shape=shape, name="bias")
return tf.Variable(initial)
# Validation set
NUM_TEST_IMAGES = 5000
print("Creating {} random test images ... ".format(NUM_TEST_IMAGES),
end="",
flush=True)
test_images, test_labels = qrcodes.getRandomBatch(size=NUM_TEST_IMAGES)
print("done")
# Inputs
with tf.name_scope("input"):
x_image = tf.placeholder(tf.float32,
shape=[None, IMAGE_SIZE, IMAGE_SIZE, 1],
name="x_image")
y_ = tf.placeholder(tf.float32, shape=[None, NUM_OUTPUTS], name="y_")
tf.summary.image('x_image', x_image, max_outputs=3)
with tf.name_scope("dropout_input"):
keep_prob = tf.placeholder(tf.float32, name="keep_probability")
# Fully connected layer
NUM_FULLY_CONNECTED_1 = 128
NUM_TRAIN_IMAGES = 500
NUM_VALID_IMAGES = 50
NUM_TEST_IMAGES = 5
#
HSIZE_IMAGE = 21
VSIZE_IMAGE = 21
#
#img = qrcode.make("asdfgjkl1234")
#img.save('./tmp.png')
#s
# Validation set
#print("Creating {} random test images ... ".format(NUM_TEST_IMAGES), end="", flush=True)
train_images, train_labels = qrcodes.getRandomBatch(size=NUM_TRAIN_IMAGES)
#print("done")
train_images = train_images.reshape(NUM_TRAIN_IMAGES, HSIZE_IMAGE, VSIZE_IMAGE)
print(train_images.shape)
#- Display Train Image sample
for i in range(3 * 4):
plt.subplot(3, 4, i + 1)
plt.imshow(train_images[i], 'gray')
plt.suptitle("train images", fontsize=12)
plt.show()
#print(string(train_labels[0]))