Add weight_column option to CSV parser #399
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
39846a3 to
cb58415
Compare
Currently the LibSVM and LibFM parsers assume 0-based indexing. However, many LibSVM files in the wild use 1-based indexing. (In fact, the [original LIBSVM implementation](https://github.com/cjlin1/libsvm/blob/master/README) uses 1-based indexing.) This PR adds a new option `indexing_mode` to both parsers to enable 1-based indexing. Description: * `indexing_mode>0`: use 1-based indexing. * `indexing_mode<0`: use 0-based indexing * `indexing_mode=0`: use heuristic to decide between the two modes of indexing. When the smallest feature index is 1 or greater, 1-based indexing is chosen. For examples, see the unit tests in `test/unittest/unittest_parser.cc`. Note. Backward compatibility is preserved as follows: when `indexing_mode` is not specified, 0-based indexing is assumed.
…to empty weight vector
7 tasks
ruslo
pushed a commit
to hunter-packages/dmlc-core
that referenced
this pull request
Mar 23, 2019
* Add weight_column option to CSV parser * Add a second test to ensure that unspecified weight_column results into empty weight vector * Also check the value field * Fix index type used in RowBlockContainer
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Outdated suggestions cannot be applied.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.
It is often useful to specify weights for training examples. Currently, it is not possible to use a separate
*.weightsfile for partitioned CSV files. To get around this, we designate a column as the weight column, from which example weights are drawn.