Sander Dieleman
Sander Dieleman
I noticed that the previous cuda-convnet result was replaced by a better one, and this time the wrapper used is pylearn2. I think both results are relevant and interesting, as...
It seems like semantic segmentation is becoming a popular task, so there are quite a few questions on the mailing list lately about doing per-pixel classification. This currently requires some...
So far the only thing that's been implemented is a bunch of tools to generate Theano expressions for neural nets. There is no actual training code in the library yet....
I think it would be more natural to specify the output width and height of a `TransformerLayer` directly, rather than specifying a downsampling factor. This is especially the case when...
This is an updated version of the paper "Discovering hidden factors of variation in deep networks" by @briancheung et al. (http://arxiv.org/abs/1412.6583), which no longer relies on any external code. The...
Here's a revamped version of the notebook I did of "Highway Networks" by Srivastava et al. (2015): http://arxiv.org/abs/1505.00387
We have a wiki page on the main library repository with a bunch of links to Lasagne extensions, as well as code that makes use of the library: https://github.com/Lasagne/Lasagne/wiki/3rd-party-extensions-and-code The...