This repository has been archived by the owner on Apr 8, 2024. It is now read-only.
Add parameter to shuffle test sets on each node during distributed training #174
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Outdated suggestions cannot be applied.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.
When using libsvm2bin component, train and test files will be converted together in bin format. It results that for distributed, you have N training files and also N testing files, and both will have to be distributed one per node. But this use case occurs mainy when you use libsvm2bin, although it can be met in other scenarios as well.
This PR implements a parameter in the training module to shuffle the test files one per node, or to share all test files across all nodes. The idea is to use mode
n_train_1_test
when using common test files, but usen_train_n_test
when the number of test files equals number of train files, and should be assigned to each node.