CNN autoencoder dimensionality reduction plus minimum-spanning-tree-based clustering for medium-scale image dataset visual exploration.
Stars
1
Forks
0
Watchers
1
Open Issues
0
Overall repository health assessment
No package.json found
This might not be a Node.js project
68
commits
Stats output now goes to a CSV file; added output.txt to the .gitignore. -Peter Mawhorter
c8b6ab1View on GitHubAdded a file referencing Mii videos, and tweaked analysis outputs a bit.
97fbf12View on GitHubUpdated output a bit for the novelty-stratified exemplars analysis;
4686db5View on GitHubAdded fractionate analysis for top 5% most-novel exemplars.
64e17d4View on GitHubMoved through various iterations of testing different
a9a0d6fView on GitHubGot rid of per-feature analysis, since it likely won't make it into the
f4410bbView on GitHubImplemented Krippendorf's alpha (although it needs debugging). Also got
cb79622View on GitHubWork on stats vs ratings, which is good, along with minor revisions to
aeff5e8View on GitHubChanged directory name for novelty examples. -Peter Mawhorter
8c47d0aView on GitHubAdded novelty example code to demonstrate what novelty "means."
bb58640View on GitHubGood amount of work with new methods in multiscale, including
bdf7d7eView on GitHub