Optimizes mean calculation routine in treeinterpreter/treeinterpreter.py #24
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
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Outdated suggestions cannot be applied.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.
I tried to use treeinterpreter to calculate feature contribution components for a large dataset which consist of 55K rows, each row ~ 15K features and even though I've parallelized my computation using Spark I was not able to run the code successfully.
One of the issues I was facing was tremendous amount of memory required by treeinterpreter for each run. It turned out that in my case most of the memory is used by _predict_forest to assemble lists of biases, contributions and predictions which are later used to calculate corresponding mean vectors.
To improve the code memory usage and runtime I propose to use iterative method for computing averages, as it is summarized in http://www.heikohoffmann.de/htmlthesis/node134.html