python - I have 8000 images, how should I turn them into features for Tensorflow? -


i'm familiar neural network architecture little rusty on image processing part.

so far, i've managed turn each 600x400 rgb image array,

array.shape = (400,600,3) 

i'm stuck @ trying create array of said arrays can scale features down.

how do that?

if want turn array in list images can feed algorithm this:

input_for_tensorflow = np.array([image1,image2,image3]) predictions = sess.run(tfpredict_tensor, feed_dict = {input_image:input_for_tensorflow}) 

in case think still lack understanding of important principles need understand create model. recommend people follow course: https://www.kadenze.com/courses/creative-applications-of-deep-learning-with-tensorflow-iv/info

hope answers question now!


Comments

Popular posts from this blog

'hasOwnProperty' in javascript -

How to put a lock and transaction on table using spring 4 or above using jdbcTemplate and annotations like @Transactional? -

How to understand 2 main() functions after using uftrace to profile the C++ program? -