Skip to content
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

Binary classification samples update #3311

Merged
merged 5 commits into from
Apr 16, 2019
Merged

Conversation

artidoro
Copy link
Contributor

Tracked in #2522.

In this PR I use the templates for binary classification samples to eliminate the dependency on the SamplesUtils and make the samples standalone.

Some binary classification trainers did not have samples, so I added them.

Note: This PR does not take care of FFM on which @wschin will be working.

@artidoro artidoro added the documentation Related to documentation of ML.NET label Apr 12, 2019
@artidoro artidoro self-assigned this Apr 12, 2019
@artidoro
Copy link
Contributor Author

artidoro commented Apr 12, 2019

Noticed that I forgot to add Gam and LinearSVM.
Gam is actually taken care by #3281. #Resolved

@shmoradims
Copy link

shmoradims commented Apr 12, 2019

using Samples.Dynamic;

this file shouldn't be in the PR. let's remove it to reduce the chance of conflicts. #Resolved


Refers to: docs/samples/Microsoft.ML.Samples/Program.cs:1 in ace5d1b. [](commit_id = ace5d1b, deletion_comment = False)

@shmoradims
Copy link

shmoradims commented Apr 12, 2019

{

all checklist points are done. please also add #6 (it's in the checklist with example links):
Inspect main properties (weights, splits, etc.) of the model. Here are examples for trees, GAM, linear. Simple is enough here. We just want to give users some ideas about what's possible. Only inspect 2-3 main model properties (not all). Also, no need to have a for-loop here. For example, if a tree model has 100 trees, take the first tree and print 2-3 main properties of that tree (NumberOfLeaves, NumberOfNodes).

When you do that might want to rename model to transformer after fitting, to avoid model.Model when you access the weights, splits, etc.

this line: var model = pipeline.Fit(trainingData);


Refers to: docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/BinaryClassification.ttinclude:11 in ace5d1b. [](commit_id = ace5d1b, deletion_comment = False)

@shmoradims
Copy link

shmoradims commented Apr 12, 2019

please also exclude libmf change in this PR #Closed

@shmoradims
Copy link

shmoradims commented Apr 12, 2019

i like this PR very much :) #Resolved

@artidoro
Copy link
Contributor Author

artidoro commented Apr 12, 2019

Add reference to:

        // This example requires installation of additional nuget package <a href="https://www.nuget.org/packages/Microsoft.ML.LightGbm/">Microsoft.ML.LightGbm</a>.

For LightGbm extensions. #Resolved

@Ivanidzo4ka
Copy link
Contributor

Ivanidzo4ka commented Apr 15, 2019

// Licensed to the .NET Foundation under one or more agreements.

What is going on with libmf folder? why you editing it? #Resolved


Refers to: src/Microsoft.ML.StandardTrainers/StandardTrainersCatalog.cs:1 in 8668701. [](commit_id = 8668701, deletion_comment = False)

@Ivanidzo4ka
Copy link
Contributor

Ivanidzo4ka commented Apr 15, 2019

        // Look at 5 predictions

Print maybe? #Resolved


Refers to: docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/BinaryClassification.ttinclude:58 in ace5d1b. [](commit_id = ace5d1b, deletion_comment = False)

@codecov
Copy link

codecov bot commented Apr 15, 2019

Codecov Report

Merging #3311 into master will decrease coverage by <.01%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #3311      +/-   ##
==========================================
- Coverage    72.7%   72.69%   -0.01%     
==========================================
  Files         807      807              
  Lines      145172   145172              
  Branches    16225    16225              
==========================================
- Hits       105541   105529      -12     
- Misses      35217    35227      +10     
- Partials     4414     4416       +2
Flag Coverage Δ
#Debug 72.69% <ø> (-0.01%) ⬇️
#production 68.22% <ø> (-0.02%) ⬇️
#test 88.97% <ø> (ø) ⬆️
Impacted Files Coverage Δ
...icrosoft.ML.Mkl.Components/MklComponentsCatalog.cs 65.71% <ø> (ø) ⬆️
...oft.ML.StandardTrainers/StandardTrainersCatalog.cs 92.34% <ø> (ø) ⬆️
...c/Microsoft.ML.FastTree/Utils/ThreadTaskManager.cs 79.48% <0%> (-20.52%) ⬇️
...soft.ML.Transforms/Text/WordEmbeddingsExtractor.cs 87.52% <0%> (-0.91%) ⬇️

@artidoro artidoro merged commit 5538ccf into dotnet:master Apr 16, 2019
@ghost ghost locked as resolved and limited conversation to collaborators Mar 22, 2022
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
documentation Related to documentation of ML.NET
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants