Skip to content

Created samples for StochasticGradientDescentNonCalibrated learner. #2770

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
Mar 2, 2019

Conversation

zeahmed
Copy link
Contributor

@zeahmed zeahmed commented Feb 27, 2019

Created samples for StochasticGradientDescentNonCalibrated learner exposed via following methods.

public static SgdNonCalibratedBinaryTrainer StochasticGradientDescentNonCalibrated(this BinaryClassificationCatalog.BinaryClassificationTrainers catalog,

public static SgdNonCalibratedBinaryTrainer StochasticGradientDescentNonCalibrated(this BinaryClassificationCatalog.BinaryClassificationTrainers catalog,

Verified

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

codecov bot commented Feb 27, 2019

Codecov Report

❗ No coverage uploaded for pull request base (master@a0edc5c). Click here to learn what that means.
The diff coverage is n/a.

@@            Coverage Diff            @@
##             master    #2770   +/-   ##
=========================================
  Coverage          ?   71.65%           
=========================================
  Files             ?      807           
  Lines             ?   142374           
  Branches          ?    16120           
=========================================
  Hits              ?   102025           
  Misses            ?    35915           
  Partials          ?     4434
Flag Coverage Δ
#Debug 71.65% <ø> (?)
#production 67.9% <ø> (?)
#test 85.84% <ø> (?)
Impacted Files Coverage Δ
...oft.ML.StandardLearners/StandardLearnersCatalog.cs 88.57% <ø> (ø)
#Resolved

@codecov
Copy link

codecov bot commented Feb 27, 2019

Codecov Report

❗ No coverage uploaded for pull request base (master@a0edc5c). Click here to learn what that means.
The diff coverage is n/a.

@@            Coverage Diff            @@
##             master    #2770   +/-   ##
=========================================
  Coverage          ?   71.68%           
=========================================
  Files             ?      808           
  Lines             ?   142392           
  Branches          ?    16112           
=========================================
  Hits              ?   102073           
  Misses            ?    35888           
  Partials          ?     4431
Flag Coverage Δ
#Debug 71.68% <ø> (?)
#production 67.92% <ø> (?)
#test 85.86% <ø> (?)
Impacted Files Coverage Δ
...oft.ML.StandardLearners/StandardLearnersCatalog.cs 88.63% <ø> (ø)

var mlContext = new MLContext(seed: 0);

// Download and featurize the dataset.
var data = SamplesUtils.DatasetUtils.LoadFeaturizedAdultDataset(mlContext);
Copy link
Member

@wschin wschin Feb 28, 2019

Choose a reason for hiding this comment

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

It'd nice if you can use in-memory data for trainers. Such a way has some benefits as discussed in #2726. #WontFix

Choose a reason for hiding this comment

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

Hi Wei-Sheng, please mark your in-memory comments as optional, until #2726 reaches a conclusion. #2726 is still in discussion and there's no clear actionable change that everybody agrees on, yet.


In reply to: 261020122 [](ancestors = 261020122)

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Lets do these things once we have decision.


In reply to: 261770052 [](ancestors = 261770052,261020122)


// Evaluate how the model is doing on the test data.
var dataWithPredictions = model.Transform(trainTestData.TestSet);
var metrics = mlContext.BinaryClassification.EvaluateNonCalibrated(dataWithPredictions);
Copy link
Member

@wschin wschin Feb 28, 2019

Choose a reason for hiding this comment

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

Need to show how user can inspect the prediction example-by-example. Evaluation is not the major part in products. #WontFix

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Lets do these things once we have decision.


In reply to: 261020336 [](ancestors = 261020336)

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:

Copy link

@shmoradims shmoradims left a comment

Choose a reason for hiding this comment

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

:shipit:

/// <example>
/// <format type="text/markdown">
/// <![CDATA[
/// [!code-csharp[StochasticGradientDescentWithOptions](~/../docs/samples/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/StochasticGradientDescentNonCalibrated.cs)]
Copy link

@shmoradims shmoradims Mar 1, 2019

Choose a reason for hiding this comment

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

StochasticGradientDescentWithOptions [](start = 26, length = 36)

please fix this copy-paste error. #Resolved

/// /// <example>
/// <format type="text/markdown">
/// <![CDATA[
/// [!code-csharp[StochasticGradientDescentWithOptions](~/../docs/samples/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/StochasticGradientDescentNonCalibratedWithOptions.cs)]
Copy link

@shmoradims shmoradims Mar 1, 2019

Choose a reason for hiding this comment

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

StochasticGradientDescentWithOptions [](start = 26, length = 36)

same here #Resolved

@zeahmed zeahmed merged commit 50474a6 into dotnet:master Mar 2, 2019
@shauheen shauheen added this to the 0319 milestone Mar 2, 2019
@ghost ghost locked as resolved and limited conversation to collaborators Mar 24, 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

5 participants