-
I notice you use batch norm both in discriminator and the generator,as shown in the code
discriminator.py:
```
def batchnorm(input):
with tf.variable_scope("batchnorm"):
# this blo…
-
std_share_network=True,
hidden_sizes=(200,200)
observed behavior: std explodes
-
### Problem
**Current defition of unknown shapes**
`shape.ndim() = 0` or any `shape[i] = 0`, for `0
-
https://arxiv.org/pdf/1506.01186.pdf
-
Hi, the results from the AdaIN implementation look amazing, and I'm looking forward to using them in some of my work. Would you be able to post a code demo in the AdaIN link: https://github.com/tensor…
-
Capturing notes from the conversation @bradenmacdonald, @symbolist, and I had with @JAAkana on the subject of course publishing granularity.
The types of publish that exist today:
* publish unit (…
-
It the latest version compatible with CUDA 10? I get the following error during the build:
```
[ 27%] Building CXX object src/CMakeFiles/mkldnn.dir/cpu/cpu_barrier.cpp.o
cd /home/mahmood/mx/mxnet…
-
It appears something in r128gain's opus handling has broken between release 0.7.1 and the latest. For example the command `r128gain -r -a -o .` previously ran without error on a directory of .opus f…
-
Hi,
I trained a fine-tuned model on my own custom dataset, and the results are generally quite accurate. However, with some images I've noticed that if I sample the same image multiple times I get …
-
When you want to train N subsequent models on N data sets, you must use --save_resume flag in the first N-1 trainings, but you SHOULD NOT use it in the last (N-th) training, if you want to get the sam…