GitHub - lorenzozanisi/pixelcnn-ood: PixelCNN for OoD detection tasks - assessing the improvement of Illustris TNG compared to Illustris

lorenzozanisi / pixelcnn-ood Public

Notifications You must be signed in to change notification settings
Fork 0
Star 2

PixelCNN for OoD detection tasks - assessing the improvement of Illustris TNG compared to Illustris

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 19 Commits
Empty_fields		Empty_fields
Illustris1		Illustris1
TNG_cutouts		TNG_cutouts
analysis		analysis
data		data
outputs		outputs
processing		processing
sersic_blobs		sersic_blobs
src		src
trained		trained
README		README

Repository files navigation

workflow is as follows:

data must be created in the Illustris1, sersic_blobs, TNG_cutouts and Empty_fields folders. The data folder also contains the script to create SDSS data (to be fixed for consistency). 

The source code for PixelCNN is in ./src

The processing (forward pass through the networks) is in ./processing . 
The dataframes for the processing can be retrieved from get_dataframes_LLR.py in this folder.

After running the processing it is necessary to run 
./analysis/likelihood_distribs.py to get the likelihood distributions.