Documentation for BinaryClassification.AveragedPerceptron #2483

shmoradims · 2019-02-08T22:33:26Z

Docs & sample for BinaryClassification.AveragedPerceptron. Related to #1209.

Ivanidzo4ka · 2019-02-08T23:01:19Z

src/Microsoft.ML.StandardLearners/StandardLearnersCatalog.cs

        /// </summary>
        /// <param name="catalog">The binary classification catalog trainer object.</param>
-        /// <param name="options">Advanced arguments to the algorithm.</param>
+        /// <param name="options">Advanced trainer options.</param>
        public static AveragedPerceptronTrainer AveragedPerceptron(


this one would love sample too!
Especially how to change loss function, and change calibrator. #Resolved

done. i don't see any public calibrator other than Platt, which is the default one. So only changed loss and a bunch of other advanced functions. Loss.Arguments() looks weird though. Users will wonder why Arguments().

In reply to: 255262343 [](ancestors = 255262343)

Ivanidzo4ka · 2019-02-08T23:02:27Z

docs/samples/Microsoft.ML.Samples/Dynamic/BinaryClassification/AveragedPerceptron.cs

+                    "occupation", "relationship", "ethnicity", "native-country", "age", "education-num", 
+                    "capital-gain", "capital-loss", "hours-per-week"))
+                // Min-max normalized all the features
+                .Append(mlContext.Transforms.Normalize("Features"))


if this is our default pipeline for binary classification can we move it to samples utils as well? I can see how it get copied again and again in other samples. #Resolved

makes sense. I'll move it as part of data loading.

In reply to: 255262549 [](ancestors = 255262549)

Ivanidzo4ka · 2019-02-08T23:03:57Z

docs/samples/Microsoft.ML.Samples/Dynamic/BinaryClassification/AveragedPerceptron.cs

+
+namespace Microsoft.ML.Samples.Dynamic.BinaryClassification
+{
+    public static class AveragedPerceptron


AveragedPerceptron [](start = 24, length = 18)

Can I convince you to move this file to Dynamic/Trainers/BinaryClassification? #Resolved

+1

In reply to: 255262779 [](ancestors = 255262779)

done

In reply to: 255266020 [](ancestors = 255266020,255262779)

Ivanidzo4ka · 2019-02-08T23:08:13Z

src/Microsoft.ML.StandardLearners/Standard/Online/OnlineLinear.cs

        [Argument(ArgumentType.AtMostOnce, HelpText = "Whether to shuffle for each training iteration", ShortName = "shuf")]
        [TlcModule.SweepableDiscreteParamAttribute("Shuffle", new object[] { false, true })]
        public bool Shuffle = true;

+        /// <summary>
+        /// Size of cache when trained in Scope.
+        /// </summary>
        [Argument(ArgumentType.AtMostOnce, HelpText = "Size of cache when trained in Scope", ShortName = "cache")]
        public int StreamingCacheSize = 1000000;



Can we delete this one? No one uses it.
It has no place in public API. #Resolved

deleted

In reply to: 255263474 [](ancestors = 255263474)

Ivanidzo4ka · 2019-02-08T23:09:57Z

src/Microsoft.ML.StandardLearners/Standard/Online/OnlineLinear.cs

        [Argument(ArgumentType.AtMostOnce, HelpText = "Initial Weights and bias, comma-separated", ShortName = "initweights")]
        [TGUI(NoSweep = true)]
        public string InitialWeights;

+        /// <summary>
+        /// Initial weights scale.
+        /// </summary>
        [Argument(ArgumentType.AtMostOnce, HelpText = "Init weights diameter", ShortName = "initwts", SortOrder = 140)]
        [TGUI(Label = "Initial Weights Scale", SuggestedSweeps = "0,0.1,0.5,1")]
        [TlcModule.SweepableFloatParamAttribute("InitWtsDiameter", 0.0f, 1.0f, numSteps: 5)]
        public float InitWtsDiameter = 0;


InitWtsDiameter [](start = 21, length = 15)

you can add this one to shortname like ShortName ="initwts, initWtsDiameter" and rename it to InitialWeightsDiameter.
#Resolved

this one is easy :)

In reply to: 255263761 [](ancestors = 255263761)

codecov · 2019-02-08T23:10:18Z

Codecov Report

❗ No coverage uploaded for pull request base (master@a254bf5). Click here to learn what that means.
The diff coverage is 0%.

@@            Coverage Diff            @@
##             master    #2483   +/-   ##
=========================================
  Coverage          ?   71.19%           
=========================================
  Files             ?      787           
  Lines             ?   140996           
  Branches          ?    16111           
=========================================
  Hits              ?   100389           
  Misses            ?    36139           
  Partials          ?     4468

Flag	Coverage Δ
#Debug	`71.19% <0%> (?)`
#production	`67.55% <0%> (?)`
#test	`85.29% <ø> (?)`

Ivanidzo4ka · 2019-02-08T23:13:14Z

src/Microsoft.ML.StandardLearners/Standard/Online/OnlineLinear.cs

        [Argument(ArgumentType.AtMostOnce, HelpText = "Number of iterations", ShortName = "iter", SortOrder = 50)]
        [TGUI(Label = "Number of Iterations", Description = "Number of training iterations through data", SuggestedSweeps = "1,10,100")]
        [TlcModule.SweepableLongParamAttribute("NumIterations", 1, 100, stepSize: 10, isLogScale: true)]
        public int NumIterations = OnlineDefaultArgs.NumIterations;

+        /// <summary>
+        /// Initial weights and bias, comma-separated.
+        /// </summary>
        [Argument(ArgumentType.AtMostOnce, HelpText = "Initial Weights and bias, comma-separated", ShortName = "initweights")]
        [TGUI(NoSweep = true)]
        public string InitialWeights;


make it internal, mark in Argument as CommandLineOnly, and create another parameter (or actually two parameters) which would accept array of floats (and single float for bias). #Resolved

you're right, but this is a non-trivial change. our current API takes this as a string. could you please file your proposal as an issue in Project 13.

In reply to: 255264280 [](ancestors = 255264280)

justinormont · 2019-02-08T23:15:42Z

docs/samples/Microsoft.ML.Samples/Dynamic/BinaryClassification/AveragedPerceptron.cs

+                // Min-max normalized all the features
+                .Append(mlContext.Transforms.Normalize("Features"))
+                // Add the trainer
+                .Append(mlContext.BinaryClassification.Trainers.AveragedPerceptron("IsOver50K", "Features"));


For the AP docs, please set iter=10. This is one of our more heavily benchmarked results; we decided to change the default iter to 10 a year or so back, though haven't thus far (see email: "Move AveragedPerceptron defaults to iter=10").

Could also go ahead and change the default: #2305 -- not docs/samples though will simplify the docs. #WontFix

Changing the default will change a whole bunch of test baselines. It needs a separate PR.

In reply to: 255264655 [](ancestors = 255264655)

Agreed; many baselines will change.

justinormont · 2019-02-08T23:18:48Z

docs/samples/Microsoft.ML.Samples/Dynamic/BinaryClassification/AveragedPerceptron.cs

+                .Append(mlContext.Transforms.Categorical.OneHotEncoding("occupation"))
+                .Append(mlContext.Transforms.Categorical.OneHotEncoding("relationship"))
+                .Append(mlContext.Transforms.Categorical.OneHotEncoding("ethnicity"))
+                .Append(mlContext.Transforms.Categorical.OneHotEncoding("native-country"))


Can OneHotEncoding() take a set of columns to transform? It might be cleaner. #Resolved

I'd like that too. The current way to transform multiple columns in one transform is to use the overload that takes an array of ColumnInfo objects. It can't be done with just strings. I think this longer version is more readable for samples. Involving ColumnInfo is a more advanced usage and is not the main focus of this sample.

If you want to see the API that you like, please add an issue for project 13.

In reply to: 255265144 [](ancestors = 255265144)

If the ColumnInfo form of the API isn't better (or good enough for docs) we should revisit the topic. I certainly see and agree to your point that the longer form is more readable. I expect the current longer form takes N-passes of the data, which is sub-optimal. Perhaps we could make a more readable multi-input API form?

@TomFinley -- thoughts? #Resolved

to be clear, I'm not saying ColumnInfo is not good for docs. I'm saying that usage is not best for this sample which is about a binary classification task. A usage of ColumnInfo will be part of samples for OneHotEncoding.

In reply to: 255756620 [](ancestors = 255756620)

Ivanidzo4ka · 2019-02-08T23:20:13Z

// TODO: Check if it works properly if Averaged is set to false

I like this line.
It smells like confidence! #Resolved

Refers to: src/Microsoft.ML.StandardLearners/Standard/Online/AveragedLinear.cs:15 in 4222e85. [](commit_id = 4222e85, deletion_comment = False)

rogancarr

First round of comments. I like the approach here. Mostly comments about the comments.

rogancarr · 2019-02-08T22:43:37Z

src/Microsoft.ML.SamplesUtils/ConsoleUtils.cs

+{
+    public static class ConsoleUtils
+    {
+        public static void PrintBinaryClassificationMetrics(BinaryClassificationMetrics metrics)


How about PrintMetrics and we overload it with the appropriate metrics type? #Resolved

Good idea!

In reply to: 255259027 [](ancestors = 255259027)

rogancarr · 2019-02-08T22:44:37Z

src/Microsoft.ML.StandardLearners/Standard/Online/AveragedLinear.cs

@@ -18,36 +18,71 @@ namespace Microsoft.ML.Trainers.Online
 {
    public abstract class AveragedLinearArguments : OnlineLinearArguments
    {
+        /// <summary>
+        /// <a href="tmpurl_lr">Learning rate</a>


In the summary, just put what it is, and add any details in Remarks. #Resolved

Remarks are only visible when the user clicks on the property. Summary is readable from the owner class. I think it's more useful to have the link here, if it doesn't make the text too detailed. Then the user will have one fewer click to do. what do you think?

In reply to: 255259207 [](ancestors = 255259207)

rogancarr · 2019-02-08T22:45:00Z

src/Microsoft.ML.StandardLearners/Standard/Online/AveragedLinear.cs

        [Argument(ArgumentType.AtMostOnce, HelpText = "Learning rate", ShortName = "lr", SortOrder = 50)]
        [TGUI(Label = "Learning rate", SuggestedSweeps = "0.01,0.1,0.5,1.0")]
        [TlcModule.SweepableDiscreteParam("LearningRate", new object[] { 0.01, 0.1, 0.5, 1.0 })]
        public float LearningRate = AveragedDefaultArgs.LearningRate;

+        /// <summary>
+        /// <see langword="true" /> to decrease the <a href="tmpurl_lr">learning rate</a> as iterations progress; otherwise, <see langword="false" />.


These seems more like Remarks. #Resolved

A comment of this form would typically be in a <value> element. #Resolved

seems to be the correct way to do these. I changed them all to tag.

In reply to: 255259283 [](ancestors = 255259283)

good point. i change them to .

In reply to: 255268625 [](ancestors = 255268625)

rogancarr · 2019-02-08T22:45:30Z

src/Microsoft.ML.StandardLearners/Standard/Online/AveragedLinear.cs

        [Argument(ArgumentType.AtMostOnce, HelpText = "Learning rate", ShortName = "lr", SortOrder = 50)]
        [TGUI(Label = "Learning rate", SuggestedSweeps = "0.01,0.1,0.5,1.0")]
        [TlcModule.SweepableDiscreteParam("LearningRate", new object[] { 0.01, 0.1, 0.5, 1.0 })]
        public float LearningRate = AveragedDefaultArgs.LearningRate;

+        /// <summary>
+        /// <see langword="true" /> to decrease the <a href="tmpurl_lr">learning rate</a> as iterations progress; otherwise, <see langword="false" />.
+        /// Default is <see langword="false" />.


This isn't necassary; this is already given in the tooltip. #Resolved

seems to be the correct way to do these. I changed them all to tag.

also tooltip is just for intellisense. it doesn't exist in docs.com

In reply to: 255259388 [](ancestors = 255259388)

It should exist in both intellisense and the docs API reference under the given class or method.

rogancarr · 2019-02-08T22:46:02Z

src/Microsoft.ML.StandardLearners/Standard/Online/AveragedLinear.cs

        [Argument(ArgumentType.AtMostOnce, HelpText = "Decrease learning rate", ShortName = "decreaselr", SortOrder = 50)]
        [TGUI(Label = "Decrease Learning Rate", Description = "Decrease learning rate as iterations progress")]
        [TlcModule.SweepableDiscreteParam("DecreaseLearningRate", new object[] { false, true })]
        public bool DecreaseLearningRate = AveragedDefaultArgs.DecreaseLearningRate;

+        /// <summary>
+        /// Number of examples after which weights will be reset to the current average.
+        /// Default is <see langword="null" />, which disables this feature.


No need to talk about defaults; The discussion can be moved to remarks. #Resolved

I don't think there's a standard way for this. I was basing my template on some existing dotnet docs below. We just need to be consistent.

https://docs.microsoft.com/en-us/dotnet/api/microsoft.office.interop.excel.defaultweboptions?view=excel-pia

https://docs.microsoft.com/en-us/dotnet/api/system.web.ui.updatepanel.childrenastriggers?view=netframework-4.7.2#property-value

In reply to: 255259488 [](ancestors = 255259488)

seems to be the correct way to do these. I changed them all to tag.

In reply to: 255683496 [](ancestors = 255683496,255259488)

rogancarr · 2019-02-08T23:14:19Z

src/Microsoft.ML.SamplesUtils/ConsoleUtils.cs

+
+namespace Microsoft.ML.SamplesUtils
+{
+    public static class ConsoleUtils


Good idea! #Resolved

rogancarr · 2019-02-08T23:16:06Z

docs/samples/Microsoft.ML.Samples/Dynamic/BinaryClassification/AveragedPerceptron.cs

+            // Create data processing pipeline
+            var pipeline =
+                // Convert categorical features to one-hot vectors
+                mlContext.Transforms.Categorical.OneHotEncoding("workclass")


The featurization here is a bit complex for discussing a learner. Can you please move this into the LoadAdultDataset method, with an optional parameter (default false) that applies this featurization? #Resolved

Can you start on the same line as pipeline? #Resolved

i want to put // comment for each section of the pipeline. . I can't add (// Convert categorical features to one-hot vectors) if I put in the same line.

In reply to: 255265735 [](ancestors = 255265735)

OneHotEncoding is the minimum we need for handling text features. we can't really use this dataset otherwise. I'm moving the pipeline to utils, but the trainer can't get an acceptable AUC by making this transform optional. agree?

In reply to: 255264730 [](ancestors = 255264730)

rogancarr · 2019-02-08T23:16:57Z

docs/samples/Microsoft.ML.Samples/Dynamic/BinaryClassification/AveragedPerceptron.cs

+                // Min-max normalized all the features
+                .Append(mlContext.Transforms.Normalize("Features"))
+                // Add the trainer
+                .Append(mlContext.BinaryClassification.Trainers.AveragedPerceptron("IsOver50K", "Features"));


Can you do a `CopyColumns("Label", "IsOver50K") and drop the parameters, using the defaults for Label and Features? #Resolved

I don't want to give the impression that the label column must be renamed to 'Label' in order for the trainer to use it. I think we do that for a lot of other examples. It's valuable for users to see that they can use the the label column as-is.

In reply to: 255264853 [](ancestors = 255264853)

rogancarr · 2019-02-08T23:23:15Z

docs/samples/Microsoft.ML.Samples/Dynamic/BinaryClassification/AveragedPerceptron.cs

+    {
+        public static void Example()
+        {
+            // In this examples we will use the adult income dataset. The goal is to predict


Can you move this to <Summary> and <Remarks> on Example? #Resolved

IMO, let's stay away from xml documentation inside samples. I don't see any of that in .Net samples, and that is harder to read that plain text.

cc @JRAlexander

In reply to: 255265822 [](ancestors = 255265822)

I checked with John. For tutorials they don't use any xml comments. Just regular // comments. we can follow the same template.

https://github.com/dotnet/samples/blob/master/machine-learning/tutorials/SentimentAnalysis/Program.cs

In reply to: 255266200 [](ancestors = 255266200,255265822)

rogancarr · 2019-02-08T23:23:31Z

docs/samples/Microsoft.ML.Samples/Dynamic/BinaryClassification/AveragedPerceptron.cs

+
+            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
+            // as a catalog of available operations and as the source of randomness.
+            // Setting the seed to a fixed number in this examples to make outputs deterministic.


example #Resolved

done

In reply to: 255265863 [](ancestors = 255265863)

sfilipi · 2019-02-08T23:28:03Z

src/Microsoft.ML.SamplesUtils/ConsoleUtils.cs

+
+namespace Microsoft.ML.SamplesUtils
+{
+    public static class ConsoleUtils


ConsoleUtils [](start = 24, length = 12)

XML docs here and in the method below. #Resolved

sfilipi · 2019-02-08T23:29:33Z

src/Microsoft.ML.StandardLearners/StandardLearnersCatalog.cs

+        /// Perceptron is a classification algorithm that makes its predictions based on a linear function.
+        /// For instance with feature values f0, f1,..., f_D-1, the prediction is given by the sign of sigma[0, D-1] (w_i * f_i), where w_0, w_1,..., w_D-1 are the weights computed by the algorithm.
+        ///
+        /// Perceptron is an online algorithm, i.e., it processes the instances in the training set one at a time.


Perc [](start = 12, length = 4)

will want to wrap in to create the paragraphs. #Resolved

<para> -- xml tag missing from comment on github page.

In reply to: 255266758 [](ancestors = 255266758)

I'm running docfx locally and the text turns out fine as is. I think the extra /// will serve as

I'll review it in smoke-test to make sure it looks ok.

In reply to: 255269520 [](ancestors = 255269520,255266758)

sfilipi · 2019-02-08T23:31:33Z

src/Microsoft.ML.StandardLearners/StandardLearnersCatalog.cs

        /// <param name="catalog">The binary classification catalog trainer object.</param>
        /// <param name="labelColumn">The name of the label column, or dependent variable.</param>
        /// <param name="featureColumn">The features, or independent variables.</param>
-        /// <param name="lossFunction">The custom loss.</param>
+        /// <param name="lossFunction">The custom <a href="tmpurl_loss">loss</a>. If <see langword="null"/>, hinge loss will be used resulting in max-margin averaged perceptron.</param>


tmpurl_loss [](start = 59, length = 11)

just a note to resolve when you put the actual URL #Resolved

yes. please see #2356

In reply to: 255267070 [](ancestors = 255267070)

sfilipi · 2019-02-08T23:33:09Z

src/Microsoft.ML.StandardLearners/Standard/Online/AveragedLinear.cs

        [Argument(ArgumentType.AtMostOnce, HelpText = "Learning rate", ShortName = "lr", SortOrder = 50)]
        [TGUI(Label = "Learning rate", SuggestedSweeps = "0.01,0.1,0.5,1.0")]
        [TlcModule.SweepableDiscreteParam("LearningRate", new object[] { 0.01, 0.1, 0.5, 1.0 })]
        public float LearningRate = AveragedDefaultArgs.LearningRate;

+        /// <summary>
+        /// <see langword="true" /> to decrease the <a href="tmpurl_lr">learning rate</a> as iterations progress; otherwise, <see langword="false" />.


[](start = 52, length = 20)

i think the can live above, and here can be a cref. #Resolved

yes. cref is better.

In reply to: 255267344 [](ancestors = 255267344)

sfilipi · 2019-02-08T23:36:16Z

src/Microsoft.ML.StandardLearners/Standard/Online/AveragedPerceptron.cs

            [Argument(ArgumentType.Multiple, HelpText = "Loss Function", ShortName = "loss", SortOrder = 50)]
            public ISupportClassificationLossFactory LossFunction = new HingeLoss.Arguments();

+            /// <summary>
+            /// The <a href="tmpurl_calib">calibrator</a> for producing probabilities. Default is exponential (aka Platt) calibration.


tmpurl_calib [](start = 29, length = 12)

one more #Resolved

these are by design: #2356

In reply to: 255267865 [](ancestors = 255267865)

…or ICanSaveModel (dotnet#2431)

…of ML.NET (dotnet#2470) Adding a project for functional tests without internal access to the ML.NET Library.

…otnet#2434) * Add analyzer for detecting BestFriend usages on public declaration * Rewrite analyzer as Symbol analyzer * Add test to prove IntClass.PubMember reports a warning

…m data. (dotnet#2475)

* removing the Microsoft.ML.Learners and the Microsoft.ML.Trainers.SymSGD namespaces. * removing the Microsoft.ML.Internal.Internallearn.ResultProcessor and the Microsoft.ML.Trainers.FastTree.Internal namespaces. * regenerating the catalog

* Reduce public API by making non-estimator components internal. * Internalize forecasting code. * Make misc stuff internal. * build break. * Make time series state internal. * Hide SequenceModelerBase * PR feedback. * PR feedback and hide more stuff. * fix entrypoint catalog test. * PR feedback. * merge conflict resolved.

* Switch OneHotEncoding parameters * Fix table of contents * Update for spacing

Ivanidzo4ka · 2019-02-11T18:01:11Z

src/Microsoft.ML.StandardLearners/Standard/Online/OnlineLinear.cs

@@ -20,24 +20,40 @@ namespace Microsoft.ML.Trainers.Online

    public abstract class OnlineLinearArguments : LearnerInputBaseWithLabel
    {
+        /// <summary>
+        /// Number of training iterations through the data.
+        /// </summary>
        [Argument(ArgumentType.AtMostOnce, HelpText = "Number of iterations", ShortName = "iter", SortOrder = 50)]
        [TGUI(Label = "Number of Iterations", Description = "Number of training iterations through data", SuggestedSweeps = "1,10,100")]
        [TlcModule.SweepableLongParamAttribute("NumIterations", 1, 100, stepSize: 10, isLogScale: true)]
        public int NumIterations = OnlineDefaultArgs.NumIterations;


NumIterations [](start = 19, length = 13)

Can we unify it across all codebase to NumberOfIterations? #Resolved

changed this occurrence. will update the other ones as I moved through the other learners.

In reply to: 255625703 [](ancestors = 255625703)

shmoradims · 2019-02-11T20:26:57Z

// TODO: Check if it works properly if Averaged is set to false

somebody has to the read the paper and see what would an AveragePerceptron even mean without averaging.

In reply to: 461980372 [](ancestors = 461980372)

Refers to: src/Microsoft.ML.StandardLearners/Standard/Online/AveragedLinear.cs:15 in 4222e85. [](commit_id = 4222e85, deletion_comment = False)

* Replace ConditionalFact usages with custom facts * Refactor ConditionalTheory usages Add AttributeUsage attributes * Make LessThanNetCore30OrNotNetCoreFactAttribute require an explicit skip message * Use correct target framework moniker for netcoreapp3.0 * Remove CORECLR constant * Remove an unused fact attribute

…pace rationalization (dotnet#2491) * moving Microsoft.ML.TimeSeriesProcessing, and Microsoft.ML.TimeSeries to Microsoft.ML.Transforms.TimeSeries * Microsoft.ML.Ensemble to Microsoft.ML.Trainers.Ensemble

…elearning into doc_binary_clf

For sample code, changed numIterations to 10 as per Justin's request

shmoradims · 2019-02-12T23:50:19Z

Please see #2517 instead. Something got messed up during rebase. I had to recreate this PR.

Shahab Moradi added 2 commits February 7, 2019 16:10

Updated docs for AveragedPerceptron

bf9d946

Added a sample

4222e85

shmoradims requested review from sfilipi and rogancarr February 8, 2019 22:33

Ivanidzo4ka reviewed Feb 8, 2019

View reviewed changes

justinormont reviewed Feb 8, 2019

View reviewed changes

Internalize and cleanup recommender project (dotnet#2451)

48dffb0

rogancarr reviewed Feb 8, 2019

View reviewed changes

sfilipi requested review from JRAlexander and luisquintanilla February 8, 2019 23:26

sfilipi reviewed Feb 8, 2019

View reviewed changes

Remove dead code (dotnet#2481)

cdc78df

sfilipi mentioned this pull request Feb 9, 2019

Lockdown of Microsoft.ML.LightGBM public surface. #2476

Merged

artidoro and others added 5 commits February 8, 2019 19:03

ITransformer derives from ICanSaveModel and explicit implementation f…

6526a01

…or ICanSaveModel (dotnet#2431)

Add a project for functional tests without visibility into internals …

7dfadca

…of ML.NET (dotnet#2470) Adding a project for functional tests without internal access to the ML.NET Library.

Add analyzer for detecting BestFriend usages on public declarations (d…

07580a8

…otnet#2434) * Add analyzer for detecting BestFriend usages on public declaration * Rewrite analyzer as Symbol analyzer * Add test to prove IntClass.PubMember reports a warning

TensorFlow: Fixed shape issue where unknown shape will be induced fro…

7d7ebb6

…m data. (dotnet#2475)

codemzs and others added 2 commits February 11, 2019 12:16

Update cookbook (dotnet#2494)

b863ac2

* Switch OneHotEncoding parameters * Fix table of contents * Update for spacing

Ivanidzo4ka reviewed Feb 11, 2019

View reviewed changes

mareklinka and others added 13 commits February 11, 2019 12:47

Towards 2326: Microsoft.ML.Ensemble and Microsoft.ML.TimeSeries names…

f269adc

…pace rationalization (dotnet#2491) * moving Microsoft.ML.TimeSeriesProcessing, and Microsoft.ML.TimeSeries to Microsoft.ML.Transforms.TimeSeries * Microsoft.ML.Ensemble to Microsoft.ML.Trainers.Ensemble

Addressed PR comments

f4f03ba

Added sample for the second overload with trainer options.

ceb3aa2

Get rid of value tuples in TrainTest and CrossValidation (dotnet#2507)

56607ba

Fixed the failing tests

2b96f6d

Updated docs for AveragedPerceptron

87be8dc

Added a sample

ba0abff

Addressed PR comments

a5538ed

Added sample for the second overload with trainer options.

6b56065

Fixed the failing tests

c487733

Merge branch 'doc_binary_clf' of https://github.com/shmoradims/machin…

5219d0c

…elearning into doc_binary_clf

Fixed breaking changes from master.

59673b8

For sample code, changed numIterations to 10 as per Justin's request

shmoradims mentioned this pull request Feb 12, 2019

Documentation for BinaryClassification.AveragedPerceptron (V2) #2517

Merged

shmoradims closed this Feb 12, 2019

shmoradims deleted the doc_binary_clf branch March 11, 2019 20:53

ghost locked as resolved and limited conversation to collaborators Mar 24, 2022

Documentation for BinaryClassification.AveragedPerceptron #2483

Documentation for BinaryClassification.AveragedPerceptron #2483

Conversation

shmoradims commented Feb 8, 2019

Ivanidzo4ka Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

shmoradims Feb 12, 2019 • edited Loading

Choose a reason for hiding this comment

Ivanidzo4ka Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Ivanidzo4ka Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Ivanidzo4ka Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Ivanidzo4ka Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

codecov bot commented Feb 8, 2019 • edited Loading

Codecov Report

Ivanidzo4ka Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

justinormont Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

justinormont Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

justinormont Feb 12, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Ivanidzo4ka commented Feb 8, 2019 • edited by shmoradims Loading

rogancarr left a comment

Choose a reason for hiding this comment

rogancarr Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

sharwell Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

rogancarr Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

rogancarr Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

sfilipi Feb 8, 2019 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rogancarr Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

sfilipi Feb 8, 2019 • edited by shmoradims Loading

Choose a reason for hiding this comment

Ivanidzo4ka Feb 8, 2019 •

edited by shmoradims

Loading

shmoradims Feb 12, 2019 •

edited

Loading

Ivanidzo4ka Feb 8, 2019 •

edited by shmoradims

Loading

Ivanidzo4ka Feb 8, 2019 •

edited by shmoradims

Loading

Ivanidzo4ka Feb 8, 2019 •

edited by shmoradims

Loading

Ivanidzo4ka Feb 8, 2019 •

edited by shmoradims

Loading

codecov bot commented Feb 8, 2019 •

edited

Loading

Ivanidzo4ka Feb 8, 2019 •

edited by shmoradims

Loading

justinormont Feb 8, 2019 •

edited by shmoradims

Loading

justinormont Feb 8, 2019 •

edited by shmoradims

Loading

justinormont Feb 12, 2019 •

edited by shmoradims

Loading

Ivanidzo4ka commented Feb 8, 2019 •

edited by shmoradims

Loading

rogancarr Feb 8, 2019 •

edited by shmoradims

Loading

rogancarr Feb 8, 2019 •

edited by shmoradims

Loading

rogancarr Feb 8, 2019 •

edited by shmoradims

Loading

sharwell Feb 8, 2019 •

edited by shmoradims

Loading

rogancarr Feb 8, 2019 •

edited by shmoradims

Loading

rogancarr Feb 8, 2019 •

edited by shmoradims

Loading

rogancarr Feb 8, 2019 •

edited by shmoradims

Loading

rogancarr Feb 8, 2019 •

edited by shmoradims

Loading

rogancarr Feb 8, 2019 •

edited by shmoradims

Loading

rogancarr Feb 8, 2019 •

edited by shmoradims

Loading

rogancarr Feb 8, 2019 •

edited by shmoradims

Loading

sfilipi Feb 8, 2019 •

edited

Loading

rogancarr Feb 8, 2019 •

edited by shmoradims

Loading

sfilipi Feb 8, 2019 •

edited by shmoradims

Loading

sfilipi Feb 8, 2019 •

edited by shmoradims

Loading

sfilipi Feb 8, 2019 •

edited by shmoradims

Loading

sfilipi Feb 8, 2019 •

edited by shmoradims

Loading

sfilipi Feb 8, 2019 •

edited by shmoradims

Loading

Ivanidzo4ka Feb 11, 2019 •

edited by shmoradims

Loading

shmoradims commented Feb 11, 2019 •

edited

Loading