The headless Notebook execution on EOXHub currently only provides limited progress information, i.e. the percentage of completed cells from the total cells in the Notebook.
In case there are only one or a limited number of cells in the Notebook that do the majority of the processing, such a percentage might not reveal a lot about the actual progress. E.g. for long running machine learning model training more details might be needed, allowing users to intervene (i.e. cancel) the processing to avoid running up expenses.
EOX indicated that papermill is used for the Notebook execution. According to their website there might be some options:
The headless Notebook execution on EOXHub currently only provides limited progress information, i.e. the percentage of completed cells from the total cells in the Notebook.
In case there are only one or a limited number of cells in the Notebook that do the majority of the processing, such a percentage might not reveal a lot about the actual progress. E.g. for long running machine learning model training more details might be needed, allowing users to intervene (i.e. cancel) the processing to avoid running up expenses.
EOX indicated that papermill is used for the Notebook execution. According to their website there might be some options:
Or perhaps other solutions can be explored.