CODEWITHSUNDEEP
pythonnumpykeras
m33n
m33n

Reputation: 1751

Keras open images and create a batch with them

I have the following function for reading images from the SO:

from keras.preprocessing import image as kimage
from keras.applications.vgg16 import preprocess_input

def read_image(path):
    img = kimage.load_img(path, target_size=(224, 224))
    tmp = kimage.img_to_array(img)
    tmp = np.expand_dims(tmp, axis=0)
    tmp = preprocess_input(tmp)
    return tmp

And I create the following data generator that basically iterates my path strings and call the previous function. I want to call read image function and stack its output into a numpy array that I can input to keras fit function. I am using this code:

batch_holder = np.zeros((batch_size, 224, 244, 3))
for j, row in batch.iterrows():
    batch_holder[j, :] = read_image(row['path'])[0]

But this does not work and I am receiving the following error which I consider as impossible to interpret:

File "train.py", line 71, in data_generator batch_holder[j, :] = read_image(row['path'])[0] ValueError: could not broadcast input array from shape (224,224,3) into shape (224,244,3)

What I am doing wrong?

Upvotes: 0

Views: 1477

Answers (1)

hugoncosta
hugoncosta

Reputation: 71

Have you written 244 instead of 224 by mistake maybe?

batch_holder = np.zeros((batch_size, 224, 244, 3))

Upvotes: 2

Related Questions