training.py: two tweaks to feature selection #226
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
1. Include posting amounts as a feature. This allows us to distinguish different classes of payments to the same payee (e.g. recurring membership fees, which often have a constant amount, from individual purchases). 2. For example key/value pairs, include the key by itself (with no substring of the value) as a feature. This is useful because different account types often have non-overlapping sets of example keys, and including the bare key as a value allows the decision tree to be effectively segmented by account type fairly close to the root. These two very small changes significantly improve training accuracy on my journal, from 94.81% to 99.32% (an 86% reduction in error rate!).
jktomer
force-pushed
the
training-tweaks
branch
from
9011bf2
to
14248ed
Compare
|
Awesome! 99% is really good. I find mine much lower. Do you have any suggestions for how to diagnose? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Include posting amounts as a feature. This allows us to distinguish different classes of payments to the same payee (e.g. recurring membership fees, which often have a constant amount, from individual purchases).
For example key/value pairs, include the key by itself (with no substring of the value) as a feature. This is useful because different account types often have non-overlapping sets of example keys, and including the bare key as a value allows the decision tree to be effectively segmented by account type fairly close to the root.
These two very small changes significantly improve training accuracy on my journal, from 94.81% to 99.32% (an 86% reduction in error rate!).