This workflow demonstrates using a Jupyter notebook from within KNIME to do a t-SNE embedding of a set of documents.

This workflow demonstrates the usage of the Python Script (DB) node that allows you to access data from a database directly within your Python script. Using this node you can alter the data within a database row by row without the need to load all the data at once.

This workflow uses a query against a SQL version of the ChEMBL database to retrieve a bunch of information about user-provided targets. It's primarily intended to be used as either a web service (deployed via the KNIME Server) or by calling it from Python (using KNIME's Python integration). In order to execute the workflow successfully, you will need to provide connection information to a database server with ChEMBL_24 installed. The "Retrieve assays, activities, and targets" wrapped metanode contains the nodes where the connection data needs to be entered.

BLOG: KNIME and Jupyter https://www.knime.com/blog/knime-and-jupyter
TAGS: Jupyter, Python, visualization, exploration, topic modeling, database, text mining, ChEMBL