This node reads word vector models of different formats:
- KNIME - Models previously saved by the Word Vector Writer Node.
- Text - Models in standard text format (file ending may be .txt or .csv). Each row contains the word in the first column and the vector in the following columns with the following properties:
Column separator: single whitespace (multiple whitespace, tab and comma are not supported)
Decimal separator: dot - Binary - Compressed or uncompressed binary model like the well known Google News Vectors model (file ending should be .bin.gz).
- Google News Vectors Note: Very large model, may take some time to read.
- GloVe Note: These models are in text format and need to be extracted first.
This node can access a variety of different file systems. More information about file handling in KNIME can be found in the official File Handling Guide.
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.