curtis.abcd
@transfluxus Hello, It is a 2d coordinates for each nodes. So, in this example, we have four destination to visit and they are located at (0, 0), (0,1), (0,2), (0,3),...
@transfluxus currently, it supports only 2d points. But you can use 'add_by_distances' function, which is not related to the number of dimensions. I guess I will update to support multi...
@lxjhk you can call add_by_distances with arguments of 2d matrix which includes distances between nodes. (This means you have to calculate distances)
@anatolix so, what did you change? Could you elaborate it easily, so that I can understand?
Yes, sure. But you will need to tune hyperparameters for each new architecture.
Hi @LinxiFan I'm also trying to reproduce the result : https://github.com/kakaobrain/fast-autoaugment/tree/dev/efficientnet I got the same result on EfficientNet-b0 after I fixed many-many things including RMSProps and EMA. (indeed, EMA caused...
@kasperpeulen I created a pull req. https://github.com/kasperpeulen/discourse-mathjax/pull/39
I Think.. It would be awesome that an user can change this url in configuration page. I'm working on this in my forked repository...
@cvasfi I think you're right. I have implemented the converted version at that paragraph, here : https://github.com/ildoonet/tf-lcnn/blob/master/layers/LookupConvolution2d.py#L196 Here is what you need to do. 1. Compile custom tf operation :...