python - Implementation of methods in Tensorflow -
i'm trying activate neural network using tensorflow tools. of way there 1 method trying implement (so can use on data) many problems. method is
from tensorflow.examples.tutorials.mnist import input_data mnist = input_data.read_data_sets(...) ... x, y = mnist.train.next_batch(batch_size)
there nice explanation using tensorflow can't find how method implemented.
every iteration need take different part data (and think method return data + y).
attached code case:
file = open('stub.csv') reader = csv.reader(file) temp = list(reader) del temp[0] ... temp in range(int(len(data) / batchsize)): ex, ey = takenextbatch(i) # takes 500 examples += 1 temp, cos = sess.run([optimizer, cost], feed_dict= {x:ex, y:ey}) # start session optimize cost function epochloss += cos
when "takenextbatch(i)" method tried write.
Comments
Post a Comment