Chiyuan Zhang
Chiyuan Zhang
@cinvro I checked the code and did not find the bug. It seems the pooling loop is not executed (otherwise `maxh` and `maxw` should not be zero). Can you at...
I think it will not show a window in command line mode, so you will need to save the figure to file to view it later on. Showing a popup...
You need to run the script in that folder to download pretrained model first.
@tenthdimension Are you running on some big endian architectures? Or maybe your downloaded data gets corrupted?
@hpoit You need to run `get-mnist.sh` which will download the MNIST data and then run convert.jl to convert them into HDF5 format.
@jfsantos Thanks! I think autoencoders, although unsupervised, are still trained with SGD, we just specify the label to be the same as the input data, and then in principle we...
@jfsantos Thanks for the details! I see, it is kind of do-able but not trivial. I need to think about this further.
@philtomson That is definitely a plan/goal, but maybe after the auto-encoders. The reason is that I do not know RNN enough to start implement them right away. But I think...
@zhongwen Thanks for the links!
We already have the ability to load caffe models, but you still need to manually translate the model definition. Automatic translation of architecture is theoretically possible but I guess might...