H2O MOJO Predictor (Word Embedding)

This node applies a word embedding MOJO to an input column.

Note: The input will not be stripped, so leading or trailing whitespaces should be removed with a String Manipulation node before prediction if desired.

Options

Word Embedding Settings

Input column

The input column either of type String or Document.
If a String is the input, the content of its cells will be treated as one word to predict, no tokenization will be performed.
If a Document is the input, all of its (already tokenized) words will be predicted. For each word an output row will be created.

Change prediction column name

Change the name of the prediction column.

Fail if a word is not in the vocabulary

If checked, the node will fail if any of the words in the input table is missing in the vocabulary of the MOJO. Otherwise, a warning will be displayed and a missing value will be the output.

Input Ports

The MOJO. Its model category must be word embedding.

Table for prediction. Missing values will be treated as NA .

Output Ports

Table containing the predicted vector.

Installation

To use this node in KNIME, install KNIME H2O Machine Learning Integration - MOJO Extension from the following update site:

A zipped version of the software site can be downloaded here. Read our FAQs to get instructions about how to install nodes from a zipped update site.

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.