keras-extras

keras-extras copied to clipboard

@kuza55 I slightly modified your function to allow for automatically using all available gpus. I also updated the api to keras 2.x, added a .gitignore file and formatted to pep8

ghego

The model expects 2 input arrays, but only received one array.

1

I have updated this code to Keras2, but met a problem. ``` with tf.device('/cpu:0'): merged = outputs_all[0] for outputs in outputs_all[1:]: print(outputs) merged.append(K.concatenate(outputs, axis=0)) print(merged) ``` print > [, ]...

LLCF

accuracy drop

4

Hi @kuza55, When I use single GPU for training, the model trains with training accuracy of 99.99%. But when i use make_parallel. The training accuracy gets stuck at 96%. Minimum...

dhimanpd

An issue about the code

2

I just use the example code in https://medium.com/@kuza55/transparent-multi-gpu-training-on-tensorflow-with-keras-8b0016fd9012 But when I changed the number of gpu(TITAN X) from 2 to 8, I found that the speed up compared with just...

LCLL

Adapted make_parallel to Keras 2.x

1

Hi @kuza55, This update to Keras 2.x works for me. Could you check the code? If you like it, you can merge it in. Cheers, Freddy

visionscaper

Ca you provide a working example of a autoencoder? I modified the keras example to use your wrapper but it failed.

alvarouc

Error with make_parallel function

3

I got the following error while trying to use make_parallel function, ``` Traceback (most recent call last): File "model_language2motion.py", line 1335, in main(parser.parse_args()) File "model_language2motion.py", line 1202, in main args.func(args)...

AgrawalAmey

cons_vae make parallel doubles output

1

as discussed this is the example from keras with multi_gpu util. autoencoder's output is double after merge (in make parallel) https://gist.github.com/varoudis/d6a71f08f3d309cc3b7583f00616d9c0

varoudis