while True:
data_helper.shuffle_train_ins()
sys.stdout.write("Epoch %d\n#################################################" % epoch)
i = 0
training_loss = 0
training_ins_sz = len(data_helper.train_ins)
offset = 0
while offset < training_ins_sz:
if i % 50 == 0:
[ _w_embedding ] = sess.run([w_embedding])
eval_ins = data_helper.get_eval_ins_embedding(_w_embedding, EMBEDDING_SIZE)
eval_res = sess.run(eval_output, feed_dict = {eval_x:eval_ins})
eval_auc(eval_res, data_helper.eval_label)
sys.stdout.write("Mini batch trained:")
sys.stdout.flush()
[label, ins] = data_helper.get_next_batch(MINI_BATCH_SIZE, offset = offset)
offset += MINI_BATCH_SIZE
_feed_dict = { y : label }
for k in range(len(x)):
_feed_dict[x[k]] = ins[k]
sys.stdout.write(
"Epoch %d\n#################################################"
% epoch)
i = 0
training_loss = 0
training_ins_sz = len(data_helper.train_ins)
offset = 0
while offset < training_ins_sz:
if i % 50 == 0:
[_w_embedding] = sess.run([w_embedding])
eval_ins = data_helper.get_eval_ins_embedding(
_w_embedding, EMBEDDING_SIZE)
eval_res = sess.run(eval_output,
feed_dict={eval_x: eval_ins})
eval_auc(eval_res, data_helper.eval_label)
sys.stdout.write("Mini batch trained:")
sys.stdout.flush()
[label, ins] = data_helper.get_next_batch(MINI_BATCH_SIZE,
offset=offset)
offset += MINI_BATCH_SIZE
_feed_dict = {y: label}
for k in range(len(x)):
