Adding a sample for LightGbm Ranking #2650
Closed
Conversation
This file contains hidden or 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
najeeb-kazmi
requested review from
Ivanidzo4ka,
shmoradims,
rogancarr and
zeahmed
Codecov Report
@@ Coverage Diff @@
## master #2650 +/- ##
==========================================
+ Coverage 71.54% 71.56% +0.02%
==========================================
Files 800 805 +5
Lines 141848 142039 +191
Branches 16119 16122 +3
==========================================
+ Hits 101482 101652 +170
- Misses 35916 35951 +35
+ Partials 4450 4436 -14
|
you don't need these. LoadFeaturizedAdultDataset already copies "IsOver50K" to "Label" (default value) so you don't need to specify this. Refers to: docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/LightGBMBinaryClassificationWithOptions.cs:25 in b572614. [](commit_id = b572614, deletion_comment = False) |
shmoradims
reviewed
Feb 20, 2019
docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/Ranking/LightGBMRanking.cs
Outdated
Show resolved
Hide resolved
shmoradims
reviewed
Feb 20, 2019
shmoradims
reviewed
Feb 20, 2019
|
|
||
| // Output: | ||
| // DCG @N: 1.38, 3.11, 4.94 | ||
| // NDCG @N: 7.13, 10.12, 12.62 |
There was a problem hiding this comment.
7.13, 10.12, 12.62 [](start = 24, length = 18)
These numbers look very low to me. i was expecting the values to be 60+.
There was a problem hiding this comment.
Yeah, looks low to me as well. Do we have a known best solution on this dataset?
We don't need a TryConvert method, we already have a TryParse, which can be used instead.
shmoradims
reviewed
Feb 20, 2019
shmoradims
reviewed
Feb 20, 2019
typo in comment: shoudl changed to should
…et#2651) * Creation of DataViewSchemaAnnotationsExtensions and movement of public facing metadata methods. * Internalize MetadataUtils and MetadataBuilder extensions, and revert members to public. * Internalize MetadataBuilderExtensions.
* Internalization and some commenting for transforms assembly. * Internalized CheckInputColumn method.
* Internalization of OneToOne and ManyToOne Column classes * PR feedback.
Ivanidzo4ka
reviewed
Feb 20, 2019
* Internalize ModelLoadContext * Internalize VersionInfo * Internalize members of ModelSaveContext while keeping the class itself public * Internalize Repository and subclasses
…alization. (dotnet#2667) * Microsoft.ML.Internal.Internallearn namespace requires certain internalization. * PR feedback. * PR feedback.
* 1st pass. builds work locally * fixes inside Ensemble, FastTree, StandardLearners, Sweeper * fixes to Data and Sweeper assemblies
* Reorder MatrixFactorizationTrainer parameters * nit * PR feedback
* Convert VectorToImageTransform to estimator/transformer * Rename Schema, Row, etc. * Code review comments, and fix baseline of EntryPointCatalog test * Code review comments * Fix EntryPointCatalog baseline
* Increase build timeout for code coverage CI. * PR feedback.
* Adding functional tests for explainability
* Stop using System.ComponentModel.Composition Replace our MEF usage, which is only used by custom mapping transforms, with the ComponentCatalog class. Fix dotnet#1595 Fix dotnet#2422 * Rename new class to CustomMappingFactory.
shmoradims
reviewed
Feb 22, 2019
docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/Ranking/LightGBMRanking.cs
Outdated
Show resolved
Hide resolved
* Use AllowSparse=false as default in TextLoader * Update entry point catelog * Make quote- default * TextLoader uses TextLoader's default settings * Address comments * tab to \t * Revert a weird change * Address comments * Reorder arguments * Polish cookbook * Reorder arguments in static TextLoader * Also change argument order in F#
This was referenced Feb 23, 2019
ghost
locked as resolved and limited conversation to collaborators
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
Fixes #2530
Fixes #776
Adding a sample for LightGbm ranking.