Add a sample for one class matrix factorization #3282

wschin · 2019-04-10T23:11:13Z

Add missing file

codecov · 2019-04-10T23:52:44Z

Codecov Report

Merging #3282 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master    #3282      +/-   ##
==========================================
+ Coverage   72.63%   72.64%   +<.01%     
==========================================
  Files         807      807              
  Lines      145129   145192      +63     
  Branches    16220    16224       +4     
==========================================
+ Hits       105413   105472      +59     
- Misses      35298    35301       +3     
- Partials     4418     4419       +1

Flag	Coverage Δ
#Debug	`72.64% <100%> (ø)`	⬆️
#production	`68.16% <ø> (-0.01%)`	⬇️
#test	`88.97% <100%> (+0.03%)`	⬆️

Impacted Files	Coverage Δ
src/Microsoft.ML.Recommender/RecommenderCatalog.cs	`70.83% <ø> (ø)`	⬆️
...ests/TrainerEstimators/MatrixFactorizationTests.cs	`97.84% <100%> (+0.43%)`	⬆️
...c/Microsoft.ML.FastTree/Utils/ThreadTaskManager.cs	`79.48% <0%> (-20.52%)`	⬇️
src/Microsoft.ML.Maml/MAML.cs	`24.75% <0%> (-1.46%)`	⬇️
src/Microsoft.ML.DataView/DataViewType.cs	`86.82% <0%> (ø)`	⬆️
src/Microsoft.ML.DataView/VectorType.cs	`89.41% <0%> (ø)`	⬆️
...soft.ML.TestFramework/DataPipe/TestDataPipeBase.cs	`74.03% <0%> (+0.33%)`	⬆️
src/Microsoft.ML.Transforms/Text/LdaTransform.cs	`89.89% <0%> (+0.62%)`	⬆️

codemzs · 2019-04-11T00:52:48Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+            // Two columns with highest predicted score to the 2nd row (indexed by 1). If we view row index as user ID and column as game ID,
+            // the following list contains the games recommended by the trained model. Note that sometime, you may want to exclude training
+            // data from your predicted results because those games were already purchased.
+            var topColumns = results.Where(element => element.MatrixRowIndex == 1).OrderByDescending(element => element.Score).Take(2);


var topColumns = results.Where(element => element.MatrixRowIndex == 1).OrderByDescending(element => element.Score).Take(2); [](start = 12, length = 123)

Can we print the output for these and put in comments? #Resolved

codemzs · 2019-04-11T00:54:58Z

src/Microsoft.ML.Recommender/RecommenderCatalog.cs

@@ -87,6 +87,7 @@ internal RecommendationTrainers(RecommendationCatalog catalog)
            /// <format type="text/markdown">
            /// <![CDATA[
            ///  [!code-csharp[MatrixFactorization](~/../docs/samples/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/Recommendation/MatrixFactorizationWithOptions.cs)]
+            ///  [!code-csharp[MatrixFactorization](~/../docs/samples/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs)]


You may want to check this won't generate too much content for the user. I had 4 links for time series but after speaking with @natke I reduced to one. #Resolved

Matrix factorization works extremely differently with different loss functions. We must have two samples.

In reply to: 274225365 [](ancestors = 274225365)

codemzs

rogancarr · 2019-04-11T18:31:00Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+    public static class OneClassMatrixFactorizationWithOptions
+    {
+        // This example shows the use of ML.NET's one-class matrix factorization module which implements
+        // Algorithm 1 in a <a href="https://www.csie.ntu.edu.tw/~cjlin/papers/one-class-mf/biased-mf-sdm-with-supp.pdf">paper</a>.


a [](start = 26, length = 1)

"Algorithm 1 in a paper" isn't very meaningful. Can you give a longer description of what it is? #Resolved

Now we tell user it's a coordinate descent method.

In reply to: 274574781 [](ancestors = 274574781)

rogancarr · 2019-04-11T18:32:29Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+            var mlContext = new MLContext(seed: 0);
+
+            // Get a small in-memory dataset.
+            GetOneClassMatrix(out List<MatrixElement> data, out List<MatrixElement> testData);


List [](start = 34, length = 4)

nit: I would use a less-specific signature than List.

rogancarr · 2019-04-11T18:33:03Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+            // Convert the in-memory matrix into an IDataView so that ML.NET components can consume it.
+            var dataView = mlContext.Data.LoadFromEnumerable(data);
+
+            // Create a matrix factorization trainer which may consume "Value" as the training label, "MatrixColumnIndex" as the


may consume [](start = 59, length = 11)

takes #Resolved

Thanks.

In reply to: 274575597 [](ancestors = 274575597)

rogancarr · 2019-04-11T18:33:22Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+
+            // Create a matrix factorization trainer which may consume "Value" as the training label, "MatrixColumnIndex" as the
+            // matrix's column index, and "MatrixRowIndex" as the matrix's row index. Here nameof(...) is used to extract field
+            // names' in MatrixElement class.


nit: Necessary? #WontFix

I'd like to be more explicit. :)

In reply to: 274575710 [](ancestors = 274575710)

rogancarr · 2019-04-11T18:33:51Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+                NumberOfThreads = 8,
+                ApproximationRank = 32,
+                Alpha = 1,
+                // The desired of unobserved values.


// The desired of unobserved values [](start = 16, length = 35)

Unclear what this means. #Resolved

New description

// The desired values of matrix elements not specified in the training set. // If the training set doesn't tell the value at the u-th row and v-th column, // its desired value would be set 0.15. In other words, this parameter determines // the value of all missing matrix elements.

In reply to: 274575901 [](ancestors = 274575901)

rogancarr · 2019-04-11T18:34:24Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+                Alpha = 1,
+                // The desired of unobserved values.
+                C = 0.15,
+                // To enable one-class matrix factorization, the following line is required.


To enable one-class matrix factorization, the following line is required. [](start = 18, length = 74)

Suggested Rephrase: This argument enables one-class matrix factorization. #Resolved

rogancarr · 2019-04-11T18:35:23Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+            var results = mlContext.Data.CreateEnumerable<MatrixElement>(prediction, false).ToList();
+            // Feed the test data into the model and then iterate through a few predictions.
+            foreach (var pred in results.Take(15))
+                Console.WriteLine($"Predicted value at row {pred.MatrixRowIndex - 1} and column {pred.MatrixColumnIndex - 1} is {pred.Score} and its expected value is {pred.Value}.");


Console.Write [](start = 16, length = 13)

Break this line; I would suggest for such long lines using old-style {0}...{1} string formatting so you can throw the arguments onto new lines. #Resolved

This is long but makes the code readable like a text message.

In reply to: 274576483 [](ancestors = 274576483)

rogancarr · 2019-04-11T18:35:53Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+            // Predicted value at row 13 and column 0 is 0.1499254 and its expected value is 0.15.
+            // Predicted value at row 14 and column 0 is 0.1499074 and its expected value is 0.15.
+            //
+            // Note: use the advanced options constructor to set the number of threads to 1 for a deterministic behavior.


// Note: [](start = 12, length = 9)

Nice touch. #Resolved

rogancarr · 2019-04-11T18:36:40Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+            //
+            // Note: use the advanced options constructor to set the number of threads to 1 for a deterministic behavior.
+
+            // Two columns with highest predicted score to the 2nd row (indexed by 1). If we view row index as user ID and column as game ID,


Two columns with highest predicted score to the 2nd row (indexed by 1). [](start = 15, length = 71)

This sentence doesn't have any context. #Resolved

rogancarr · 2019-04-11T18:37:12Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+            //
+            // Note: use the advanced options constructor to set the number of threads to 1 for a deterministic behavior.
+
+            // Two columns with highest predicted score to the 2nd row (indexed by 1). If we view row index as user ID and column as game ID,


we [](start = 90, length = 2)

If we wanted a model to recommend video games to a user, we could view...

rogancarr · 2019-04-11T18:37:33Z

...crosoft.ML.Samples/Dynamic/Trainers/Recommendation/OneClassMatrixFactorizationWithOptions.cs

+
+            // Two columns with highest predicted score to the 2nd row (indexed by 1). If we view row index as user ID and column as game ID,
+            // the following list contains the games recommended by the trained model. Note that sometime, you may want to exclude training
+            // data from your predicted results because those games were already purchased.


those [](start = 56, length = 5)

would represent games that were #Resolved

rogancarr

Looks good! Just a few nits.

wschin requested review from codemzs, Ivanidzo4ka and sfilipi April 10, 2019 23:11

wschin self-assigned this Apr 10, 2019

Add a sample for OCMF

b2cdfb1

Add missing file

wschin force-pushed the ocmf-sample branch from 6c86c73 to b2cdfb1 Compare April 10, 2019 23:13

codemzs reviewed Apr 11, 2019

View reviewed changes

codemzs approved these changes Apr 11, 2019

View reviewed changes

wschin added 2 commits April 11, 2019 10:26

Address comments

b83ffbe

Merge branch 'master' into ocmf-sample

11478c5

rogancarr reviewed Apr 11, 2019

View reviewed changes

rogancarr approved these changes Apr 11, 2019

View reviewed changes

Address comments

a317759

wschin merged commit 89a1fb9 into dotnet:master Apr 12, 2019

wschin deleted the ocmf-sample branch April 12, 2019 00:01

ghost locked as resolved and limited conversation to collaborators Mar 22, 2022

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add a sample for one class matrix factorization #3282

Add a sample for one class matrix factorization #3282

wschin commented Apr 10, 2019

codecov bot commented Apr 10, 2019 •

edited

Loading

codemzs Apr 11, 2019 •

edited by wschin

Loading

codemzs Apr 11, 2019 •

edited by wschin

Loading

wschin Apr 11, 2019

codemzs left a comment

rogancarr Apr 11, 2019 •

edited by wschin

Loading

wschin Apr 11, 2019

rogancarr Apr 11, 2019

rogancarr Apr 11, 2019 •

edited by wschin

Loading

wschin Apr 11, 2019

rogancarr Apr 11, 2019 •

edited by wschin

Loading

wschin Apr 11, 2019

rogancarr Apr 11, 2019 •

edited by wschin

Loading

wschin Apr 11, 2019

rogancarr Apr 11, 2019 •

edited by wschin

Loading

rogancarr Apr 11, 2019 •

edited by wschin

Loading

wschin Apr 11, 2019

rogancarr Apr 11, 2019 •

edited

Loading

rogancarr Apr 11, 2019 •

edited by wschin

Loading

rogancarr Apr 11, 2019

rogancarr Apr 11, 2019 •

edited by wschin

Loading

rogancarr left a comment •

edited

Loading

Add a sample for one class matrix factorization #3282

Add a sample for one class matrix factorization #3282

Conversation

wschin commented Apr 10, 2019

codecov bot commented Apr 10, 2019 • edited Loading

Codecov Report

codemzs Apr 11, 2019 • edited by wschin Loading

Choose a reason for hiding this comment

codemzs Apr 11, 2019 • edited by wschin Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

codemzs left a comment

Choose a reason for hiding this comment

rogancarr Apr 11, 2019 • edited by wschin Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Apr 11, 2019 • edited by wschin Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Apr 11, 2019 • edited by wschin Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Apr 11, 2019 • edited by wschin Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Apr 11, 2019 • edited by wschin Loading

Choose a reason for hiding this comment

rogancarr Apr 11, 2019 • edited by wschin Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Apr 11, 2019 • edited Loading

Choose a reason for hiding this comment

rogancarr Apr 11, 2019 • edited by wschin Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Apr 11, 2019 • edited by wschin Loading

Choose a reason for hiding this comment

rogancarr left a comment • edited Loading

Choose a reason for hiding this comment

codecov bot commented Apr 10, 2019 •

edited

Loading

codemzs Apr 11, 2019 •

edited by wschin

Loading

codemzs Apr 11, 2019 •

edited by wschin

Loading

rogancarr Apr 11, 2019 •

edited by wschin

Loading

rogancarr Apr 11, 2019 •

edited by wschin

Loading

rogancarr Apr 11, 2019 •

edited by wschin

Loading

rogancarr Apr 11, 2019 •

edited by wschin

Loading

rogancarr Apr 11, 2019 •

edited by wschin

Loading

rogancarr Apr 11, 2019 •

edited by wschin

Loading

rogancarr Apr 11, 2019 •

edited

Loading

rogancarr Apr 11, 2019 •

edited by wschin

Loading

rogancarr Apr 11, 2019 •

edited by wschin

Loading

rogancarr left a comment •

edited

Loading