Skip to content

Fix iris.txt dataset and modify Iris Classification tests accordingly #428

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jul 12, 2018

Conversation

bojanmisic
Copy link
Contributor

Related to issue #400.

  1. Reordered column values in iris.txt (headers stay the same), so the previously Petal width values are actually Sepal length values; other columns shifted to the right.

  2. Tests TrainAndPredictIrisModelTest and TrainAndPredictIrisModelWithStringLabelTest are modified accordingly so that the values that test the prediction are also reordered to be more realistic (same way the dataset is modified).

  3. Model class IrisDataWithStringLabel fields reordered to match iris.data dataset.

The fix for the samples repo that uses the same dataset will follow.

@dnfclas
Copy link

dnfclas commented Jun 27, 2018

CLA assistant check
All CLA requirements met.

@bojanmisic bojanmisic changed the title Fix iris.txt dataset and modify Iris Classification tests accordingly WIP: Fix iris.txt dataset and modify Iris Classification tests accordingly Jun 27, 2018
@bojanmisic
Copy link
Contributor Author

bojanmisic commented Jun 27, 2018

Ok, just realized some additional tests need to be covered. #Resolved

@OliaG OliaG self-requested a review June 27, 2018 13:58
@sfilipi
Copy link
Member

sfilipi commented Jun 27, 2018

Thanks for taking care of this, Bojan.

Take a look at the results of the tests as well, they might need update since the dataset is changing quite a bit.


In reply to: 400616611 [](ancestors = 400616611)

@sfilipi
Copy link
Member

sfilipi commented Jun 27, 2018

        Assert.Equal(49, matrix["2", "2"]);

see if anything here need be updated.

It might be the case for other tests using this dataset. #Resolved


Refers to: test/Microsoft.ML.Tests/Scenarios/IrisPlantClassificationTests.cs:113 in 57b9fc3. [](commit_id = 57b9fc3, deletion_comment = False)

Bojan Misic added 2 commits June 28, 2018 09:13

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
@bojanmisic
Copy link
Contributor Author

Thanks Senja!

Removing WIP and adding new change:

  1. Tests MultiClassifierLightGMBKeyLabelTest and MultiClassifierLightGBMKeyLabelU404Test were failing since used Iris dataset was changed significantly, so baseline test files needed to be updated (latest commit). Pinging @codemzs since he was the author.

@bojanmisic bojanmisic changed the title WIP: Fix iris.txt dataset and modify Iris Classification tests accordingly Fix iris.txt dataset and modify Iris Classification tests accordingly Jun 28, 2018
Copy link
Contributor

@Ivanidzo4ka Ivanidzo4ka left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

:shipit:

Copy link
Member

@sfilipi sfilipi left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

:shipit:

@Ivanidzo4ka Ivanidzo4ka merged commit 54596ac into dotnet:master Jul 12, 2018
eerhardt pushed a commit to eerhardt/machinelearning that referenced this pull request Jul 27, 2018
…dotnet#428)

* Fix iris.txt dataset and modify Iris Classification tests accordingly

* Modify baseline test files for multiclass classification with Iris dataset and LightGBM
@ghost ghost locked as resolved and limited conversation to collaborators Mar 30, 2022
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

4 participants