Skip to content

Latest commit

 

History

History
85 lines (53 loc) · 5.01 KB

pca.md

File metadata and controls

85 lines (53 loc) · 5.01 KB
Raw

PCA Tutorial

This tutorial walks new users through the process of creating a Principal Components Analysis (PCA) model.

Those who have never used H2O before should refer to Getting Started for additional instructions on how to run H2O Flow.

For more details on the math behind H2O's implementation of PCA, refer to PCA Data Science.

When to Use PCA

PCA is used to reduce dimensions and solve issues of multicollinearity in high dimension data.

Getting Started

This tutorial uses a publicly available data set that can be found at: http://archive.ics.uci.edu/ml/datasets/Arrhythmia.

The original data are the Arrhythmia data set made available by UCI Machine Learning Repository. They are composed of 452 observations and 279 attributes.

To further explore H2O's capabilities, some publicly available data sets can be found on our website.

####Importing Data Before creating a model, import data into H2O:

  1. Click the Assist Me! button in the Help tab in the sidebar on the right side of the page.

Assist Me button 0. Click the importFiles link and enter the file path to the dataset in the Search entry field, or drag and drop the file onto the Search entry field and press Enter to confirm the drop. 0. Click the Add all link to add the file to the import queue, then click the Import button. Importing Files

####Parsing Data Now, parse the imported data:

  1. Click the Parse these files... button. Note: The default options typically do not need to be changed unless the data does not parse correctly.
  2. From the drop-down Parser list, select the file type of the data set (Auto, XLS, CSV, or SVMLight).
  3. If the data uses a separator, select it from the drop-down Separator list.
  4. If the data uses a column header as the first row, select the First row contains column names radio button. If the first row contains data, select the First row contains data radio button. You can also select the Auto radio button to have H2O automatically determine if the first row of the dataset contains the column names or data.
  5. If the data uses apostrophes ( ' - also known as single quotes), check the Enable single quotes as a field quotation character checkbox.
  6. To delete the imported dataset after the parse is complete, check the Delete on done checkbox.

NOTE: In general, we recommend enabling this option. Retaining data requires memory resources, but does not aid in modeling because unparsed data can’t be used by H2O.

  1. Review the data in the Data Preview section, then click the Parse button.

Parsing Data

NOTE: Make sure the parse is complete by clicking the View Job button and confirming progress is 100% before continuing to the next step, model building. For small datasets, this should only take a few seconds, but larger datasets take longer to parse.

Building a Model

  1. Once data are parsed, click the View button, then click the Build Model button.

  2. Select Principal Component Analysis from the drop-down Select an algorithm menu, then click the Build model button.

  3. If the parsed arrhythmia.hex file is not already listed in the Training_frame drop-down list, select it. Otherwise, continue to the next step.

  4. From the Ignored_columns section, select the columns to ignore in the Available area to move them to the Selected area. For this example, do not select any columns.

  5. In the K field, specify the number of clusters. For this example, enter 3.

  6. In the Max_iterations field, specify the maximum number of iterations. For this example, enter 100.

  7. From the drop-down Init menu, select the initialization mode. For this example, select PlusPlus.

    • PlusPlus initialization chooses one initial center at random and weights the random selection of subsequent centers so that points furthest from the first center are more likely to be chosen.

    • User initialization requires the corresponding User_points parameter. To define a specific initialization point, select the imported dataset .hex file from the drop-down User_points list, then select User from the drop-down Init list.

      Note: The user-specified points dataset must have the same number of columns as the training dataset.

  8. Click the Build Model button.

Building PCA Models

PCA Results

The output for PCA includes the following:

  • Output (model category, model summary, scoring history, training metrics, validation metrics, iterations)
  • Archetypes
  • Standard deviation
  • Rotation
  • Importance of components