Using nestor-dash. Visualizations don't work

[ns-100]

After uploading a HDF5 file, when I click on the visualizations I get the following error message builtins.IndexError IndexError: list index out of range

Below is the full error trace:

[rtbs-dev]

Apologies for the delay on this, many of us were out for a conference.

Interesting, seems to be requesting the feature names but none exist. Have you followed the documentation to add correct meta tags to use the dashboard feature? {docs here](https://nestor.readthedocs.io/en/latest/how_to_guide/dash_demonstration.html#uploading-data-files)

Dash is highly experimental right now, so it may not work with all datasets.

[ns-100]

I haven't fed the gui my dataset yet. I wanted to get a feel for how the tool works so used the excavator.csv dataset and it resulted in the error message above.

[rtbs-dev]

Not quite what I meant. The dashboard expects an annotated .h5 file (HDFS) to be uploaded.

• .h5, specifically, as output by the Nestor GUI (or core functions within python) upon performing annotation
• annotated, meaning that in either the initial load-up of the dataset or under Edit > Select Columns for Analysis, you must have marked at least one column as containing either Maintenance Technician and/or one as Asset ID, prior to exporting the .h5. (Note that these do not need to be "checked" as containing natural language for tagging, just as being some known "category" from the drop-down).

Unfortunately this is not very well documented yet (see here), due to the very alpha nature of the dashboard. We recommend using python/matlab/R/excel to analyse the h5 or csv output from Nestor GUI for now, if getting analysis out quickly matters to you. Otherwise we're very happy to have help testing the dashboard!

[ns-100]

I meant to add above that the .h5 that resulted from tagging the excavator.csv in the gui was used for the dashboard application and yes the Asset ID column is part of the initial load-up of the dataset. I did not mark the other csv columns to a "known" category.

I do plan to use python to supplement the analytics from the dashboard, but for now it would be great to get the .h5 to work with the dashboard to use for a prototype/poc demo.