[SPARK-8517][ML][DOC] Reorganizes the spark.ml user guide

[thunterdb]

This PR moves pieces of the spark.ml user guide to reflect suggestions in SPARK-8517. It does not introduce new content, as requested.

[jkbradley]

I'd use the "[ML]" tag in the PR title.

[jkbradley]

Remove the empty "ml-pipelines.md" file?

[SparkQA]

Test build #47358 has finished for PR 10207 at commit 451b773.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * [OneVsRest](http://en.wikipedia.org/wiki/Multiclass_classification#One-vs.-rest) is an example of a machine learning reduction for performing multiclass classification given a base classifier that can perform binary classification efficiently. It is also known as \"One-vs-All.\"\n * [Iris dataset](http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass/iris.scale), parse it as a DataFrame and perform multiclass classification usingOneVsRest. The test error is calculated to measure the algorithm accuracy.\n * The Pipelines API for Decision Trees offers a bit more functionality than the original API. In particular, for classification, users can get the predicted probability of each class (a.k.a. class conditional probabilities).\n * * a bit more functionality for random forests: estimates of feature importance, as well as the predicted probability of each class (a.k.a. class conditional probabilities) for classification.\n * public class Document implements Serializable\n * public class LabeledDocument extends Document implements Serializable\n * public class Document implements Serializable\n * public class LabeledDocument extends Document implements Serializable\n

[jkbradley]

I like using the keywords "extraction, transformation, and selection" since users may search for those. "Building" is pretty generic.

[thunterdb]

Done

[jkbradley]

Remove empty "ml-examples.md" file?

[jkbradley]

Those are my high-level comments. Have you rewritten much text? If so, I can do a second more detailed pass after updates (which will require restructuring).

Thanks!

[thunterdb]

@jkbradley no this PR just moves the text around, with little modification. More substantital changes will be done later.

[jkbradley]

OK thanks just confirming

[SparkQA]

Test build #47367 has finished for PR 10207 at commit 6c7850b.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * [OneVsRest](http://en.wikipedia.org/wiki/Multiclass_classification#One-vs.-rest) is an example of a machine learning reduction for performing multiclass classification given a base classifier that can perform binary classification efficiently. It is also known as \"One-vs-All.\"\n * [Iris dataset](http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass/iris.scale), parse it as a DataFrame and perform multiclass classification usingOneVsRest. The test error is calculated to measure the algorithm accuracy.\n * The Pipelines API for Decision Trees offers a bit more functionality than the original API. In particular, for classification, users can get the predicted probability of each class (a.k.a. class conditional probabilities).\n * * a bit more functionality for random forests: estimates of feature importance, as well as the predicted probability of each class (a.k.a. class conditional probabilities) for classification.\n * public class Document implements Serializable\n * public class LabeledDocument extends Document implements Serializable\n * public class Document implements Serializable\n * public class LabeledDocument extends Document implements Serializable\n

[SparkQA]

Test build #47369 has finished for PR 10207 at commit 216acd3.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * [OneVsRest](http://en.wikipedia.org/wiki/Multiclass_classification#One-vs.-rest) is an example of a machine learning reduction for performing multiclass classification given a base classifier that can perform binary classification efficiently. It is also known as \"One-vs-All.\"\n * [Iris dataset](http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass/iris.scale), parse it as a DataFrame and perform multiclass classification usingOneVsRest. The test error is calculated to measure the algorithm accuracy.\n * The Pipelines API for Decision Trees offers a bit more functionality than the original API. In particular, for classification, users can get the predicted probability of each class (a.k.a. class conditional probabilities).\n * * a bit more functionality for random forests: estimates of feature importance, as well as the predicted probability of each class (a.k.a. class conditional probabilities) for classification.\n * public class Document implements Serializable\n * public class LabeledDocument extends Document implements Serializable\n * public class Document implements Serializable\n * public class LabeledDocument extends Document implements Serializable\n

[jkbradley]

Use "Example" heading

[jkbradley]

Thanks! just a few comments

[jkbradley]

Thanks for updating it! LGTM pending tests.

[SparkQA]

Test build #2187 has finished for PR 10207 at commit dc584b2.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * [OneVsRest](http://en.wikipedia.org/wiki/Multiclass_classification#One-vs.-rest) is an example of a machine learning reduction for performing multiclass classification given a base classifier that can perform binary classification efficiently. It is also known as \"One-vs-All.\"\n * [Iris dataset](http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass/iris.scale), parse it as a DataFrame and perform multiclass classification usingOneVsRest. The test error is calculated to measure the algorithm accuracy.\n * The Pipelines API for Decision Trees offers a bit more functionality than the original API. In particular, for classification, users can get the predicted probability of each class (a.k.a. class conditional probabilities).\n * * a bit more functionality for random forests: estimates of feature importance, as well as the predicted probability of each class (a.k.a. class conditional probabilities) for classification.\n * public class Document implements Serializable\n * public class LabeledDocument extends Document implements Serializable\n * public class Document implements Serializable\n * public class LabeledDocument extends Document implements Serializable\n

[SparkQA]

Test build #47384 has finished for PR 10207 at commit dc584b2.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * [OneVsRest](http://en.wikipedia.org/wiki/Multiclass_classification#One-vs.-rest) is an example of a machine learning reduction for performing multiclass classification given a base classifier that can perform binary classification efficiently. It is also known as \"One-vs-All.\"\n * [Iris dataset](http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass/iris.scale), parse it as a DataFrame and perform multiclass classification usingOneVsRest. The test error is calculated to measure the algorithm accuracy.\n * The Pipelines API for Decision Trees offers a bit more functionality than the original API. In particular, for classification, users can get the predicted probability of each class (a.k.a. class conditional probabilities).\n * * a bit more functionality for random forests: estimates of feature importance, as well as the predicted probability of each class (a.k.a. class conditional probabilities) for classification.\n * public class Document implements Serializable\n * public class LabeledDocument extends Document implements Serializable\n * public class Document implements Serializable\n * public class LabeledDocument extends Document implements Serializable\n

[jkbradley]

Merging with master and branch-1.6