Generate Text Using a Many-To-One LSTM Network (Deployment)

The workflows generates text in fairy tale style.
In the “Extract Index” metanode we use the probability distribution over all possible indexes to make the predictions.
Here we have two options. We can either always predict the index with the highest probability (Deployment Workflow I) or we can pick the next index based on the given probability distribution (Deployment Workflow II) .
The last node, named Extract Predicted Text translates the sequence of indexes into characters.

In order to run the example, please make sure you have the following KNIME extensions installed:

* KNIME Deep Learning - Keras Integration
* KNIME Deep Learning - TensorFlow Integration
* KNIME Python Integration

You also need a local Python installation that includes Keras. Please refer to https://www.knime.com/deeplearning#keras for installation recommendations and further information.

External resources

• “Once Upon A Time … “ by LSTM Network Blogpost