/galaxyzoo

Entry for GalaxyZoo challenge
  1. Lua 98.5%
  2. Shell 1.5%
Lua Shell
Find file
Latest commit 7e8ece3 Nov 2, 2014 @soumith update readme
Permalink
Failed to load latest commit information.
test moving to SpatialConvolutionMM and removing the whole transpose stuff Nov 1, 2014
.gitignore data loading and partial model files Mar 21, 2014
1_data.lua moving to SpatialConvolutionMM and removing the whole transpose stuff Nov 1, 2014
1_datafunctions.lua moving to SpatialConvolutionMM and removing the whole transpose stuff Nov 1, 2014
2_model.lua moving to SpatialConvolutionMM and removing the whole transpose stuff Nov 1, 2014
3_train.lua fixed small bug in rMSE reporting Mar 29, 2014
4_test.lua moving to SpatialConvolutionMM and removing the whole transpose stuff Nov 1, 2014
INSTALL.md Update INSTALL.md May 1, 2014
LICENSE Update LICENSE Mar 20, 2014
README.md update readme Nov 1, 2014
dataprep.sh checked that the test scripts are working Mar 31, 2014
doall.lua
torchenv.rc

README.md

galaxyzoo

Entry for GalaxyZoo Challenge: http://www.kaggle.com/c/galaxy-zoo-the-galaxy-challenge

Our best score was 0.08246 placing us at 9th position, we trained the network for about 6 days, 4 models with different random seeds were trained on 2 NVidia Titan cards. The model that was trained for entry is in 2_model.lua

The winning model was Sander Dieleman's convnet appropach described here: http://benanne.github.io/2014/04/05/galaxy-zoo.html

For install instructions, please look at INSTALL.md