-
- Description:
Peewit 0.2 is a prototype implementation of a framework that is intended to support programming, running and result examination of machine learning experiments. It does not provide any sort solutions but also makes little assumptions on the sort of experiments the user wants to accomplish. Can it be of any use then?
That is the question we want to pursue with peewit. The framework is based on two observations:
1) Machine learning experiment often have some sort of uniform tree structure.
2) Experimentors happen to puzzle about which numbers belong to what parameters the next day.
Peewit is designated to experiments that fulfill a certain uniformity in the relation of the experimental components. It demands from the user to term things and rewards this by an increased live-time of names.
- Changes to previous version:
self-referential descent inputs
Comments
No one has posted any comments yet. Perhaps you'd like to be the first?
Leave a comment
You must be logged in to post comments.