In this workflow we have taken grocery dataset for 10 stores contains data for 50 different items. This workflow can be deployed for a guided analytics webportal, i.e. easy to use dashborad for time series inspecction and forecasting.
Steps:
* Data preparation: data loading, data exploration, data imputation
* Selecting parameters and filtering data: selcting particular store, item , setting parameters for weekly/monthly stats visualizations
* data visualizations: data inspection , weekly/monthly sales
* seasonality inspection: seasionality inspection, seasonality removing, lag preparation
* model training and evaluation: traing model on time series data, matics evaluation, line plot for performance
seems like statsmodels is not installed in the Python environment you use for this node.
I suggest going to Preferences -> KNIME -> Python (legacy), have a look at the name of conda environment set in the Python 3 (Default) section. Let’s say it is called py3_knime_environment. Now open a terminal, and enter according to the documentation of the node, which can be accessed via the node description:
I spent several hours in a circular loop trying to figure this out. Not sure what was different (ended up reinstalling 5.4), but now it’s successful. Much appreciated!
I don’t see a button for ‘solution’ - but this was the solution
