3.8 Linear Regression Limits

There are some limits that apply for the creation of any BigML resource. These are limits based on the number of classes, terms and items that can be considered to create your models. This is explained in Field Limits .

Additionally, some specific limits apply for your linear regressions visualization, i.e. to the linear regression chart and the coefficient table views, depending on the number of input fields in your dataset. See subsection 3.8.2 and subsection 3.8.1 for a detailed explanation.

Note: chart limits and coefficient table limits just affect to the visualization of the model, i.e., despite your dataset reach those limits, you can still creating the linear regression, evaluating it and using it to make predictions.

Field Limits

Linear regression, similarly to other BigML models, has the following limitations according to the type of field:

• Classes: for categorical, a maximum number of 1,000 distinct classes per field is allowed.

• Terms: BigML can handle up to 1,000 terms in total. If multiple text fields are defined, then the token limit per field is evenly divided by the number of text fields evenly, e.g., a dataset with two text fields would result in 500 terms per text field. BigML selects those terms with most significant frequency, discarding both those that appear either too often or too infrequently. A maximum of 256 characters per term is allowed.

• Items: a maximum number of 10,000 distinct items per field is allowed.

3.8.1 Chart Limits

There are some circumstances under which your chart cannot be displayed:

• As the 1D chart only supports numeric fields for the x-axis, if your linear regression only contains categorical, text, or items fields, the 1D chart cannot be displayed. When you try to click on the 1D chart icon you will see a warning message. (See Figure 3.102 )

  

  If your linear regression contains more than 800 input fields the chart cannot be displayed and the default view will be the table view. When you try to click on the chart icon, you will see the warning message shown in Figure 3.103 .

  

• If your linear regression contains more than 800 predictors, the prediction interval bounds cannot be displayed in the chart. You will see the warning message shown in Figure 3.103 . You can still see your coefficients by downloading the CSV file. For how to calculate the number of predictors in a linear regression, see subsection 3.2.3 .

  

3.8.2 Coefficient Table Limits

If your linear regression contains more than 1,000 input fields, the coefficient table cannot be displayed. You will need to download the CSV if you want to see your linear regression coefficients. You will get the message shown in Figure 3.105 :