Word2Vec Learner (legacy)

This node trains a Word2Vec model on unlabelled documents. The model will only learn word representations (vector for each word), that can be extracted using the Vocabulary Extractor Node. For more information on Word Vectors in general see: http://deeplearning4j.org/word2vec

The KNIME Deeplearning4J Integration has been marked as legacy with KNIME Analytics Platform 5.0 and will be deprecated in a future version. If you are using this extension in a production workflow, consider switching to one of the other deep learning integrations available in KNIME Analytics Platform.

Options

Data Options

Document Column
Column containing the Document or String to train on.

Learning Options

Learning Rate
The starting learning rate.
Minimum Learning Rate
The minimum learning rate threshold. The learning rate will decay automatically over time.
Layer Size
The length of the resulting word vectors.
Batch Size
The number of words to use for each batch.
Seed
A seed value to use for training.
Number of Epochs
The number of epochs to train.
Number of Training Iterations
The number of updates done for each batch.
Context Window Size
Size of the context, meaning a window around each word to consider for learning.
Minimum Word Frequency
Minimum frequency of a word to appear in the corpus to be considered for learning. Words with a lower frequency will not appear in the vocabulary contained in the Word Vector Model.
Sampling Rate
Threshold for configuring which higher-frequency words are randomly downsampled; a useful range is (0, 1e-5).
Use Hierarchical Softmax?
Whether to use hierarchical softmax. If checked negative sampling will be disabled.
Negative Sampling Rate
The number of “noise words” that should be drawn.
Elements Learning Algorithm
The algorithm to use to learn word representations.
Skip missing cells?
Whether rows containing missing cells should be skipped or not. If missing cells should not be skipped but the table contains missing cells the node will fail.

Input Ports

Icon
Table containing Document or String column.

Output Ports

Icon
Trained Doc2Vec Model.

Popular Predecessors

Views

This node has no views

Workflows

Links

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.