-
## 🐛 Bug
Hello, I’m running a GAN model with some second-order regularization. The training went well at the beginning and the loss decreased. However, after a random period of time, ranging from 3…
-
In the paper 《DiCENet: Dimension-wise Convolutions for Efficient Networks》, the network width scaling parameter **s** can be selected, but in the experiment of **image multi-label classification**, th…
-
------------------------
### System information
- **Have I written custom code (as opposed to using a stock example script provided in TensorFlow)**:
I used stock example script provided by Te…
-
Great list! Feel free to add these two papers related to capsule networks.
1) Graph Capsule Convolutional Neural Networks. Saurabh Verma, Zhi-Li Zhang (ICML workshop 2018, our paper :))
2) Capsul…
-
https://github.com/tensorflow/tensorflow/issues/1300 added support for fp16 storage, but there is currently no support for native fp16 computation, which is available on some hardware such as Pascal G…
-
Theano has supported multiple GPUs since v0.8.0.
(cf. [Using multiple GPUs — Theano 0.8.0 documentation](http://deeplearning.net/software/theano/tutorial/using_multi_gpu.html) )
Does Keras also suppor…
-
Recently, my supervisor encouraged me to work on graph conv in VQA, however, I found that such a graph conv model seems less powerful than other models in VQA, so if you are willing to share, could yo…
-
UPDATE: bug was found to be related to some problem with `self.datas` in `BTgymMultiData`. reason is still unclear. but issue is unrelated to filters as originally proposed
I'm currently experimen…
-
I wanted to know what are the input & output nodes of the graph generated in your code. Could you please provide me this information?
Thank you in advance
-
My recent experimentation with BTGym involve using long 'time_dim' and long episodes respectively.
in those settings both cpu usage and memory usage increase drastically.
So what I'm trying to und…