OHDSI / CohortDiagnostics

An R package for performing various cohort diagnostics.
https://ohdsi.github.io/CohortDiagnostics
41 stars 48 forks source link

Shiny not loading, warning message: Warning: Error in dplyr::bind_rows: Argument 1 must have names. #1000

Closed christinaread closed 1 year ago

christinaread commented 1 year ago

Hello,

I am experiencing an issue running the Shiny viewer. When running any of the codes: "shiny::runApp()" ; "CohortDiagnostics::launchDiagnosticsExplorer( sqliteDbPath = "data/MergedCohortDiagnosticsData_default.sqlite")", or clicking run app in the global.r file, the shiny does not load correctly. The webpage will load but not generate any information. The following warning message appears, but there is no error and the code continues to run (without producing results) indefinitely. Thank you for your help.

Listening on http://127.0.0.1:4654 Warning: Error in dplyr::bind_rows: Argument 1 must have names.
74: Warning: Error in dplyr::bind_rows: Argument 1 must have names. 74:

azimov commented 1 year ago

Hi @christinaread I will need a bit more information to debug your issue.

Firstly, which version of CohortDiagnostics are you using?

Secondly, is the path data/MergedCohortDiagnosticsData_default.sqlite the correct location of your output file?

Thirdly, are you sure you generated the merged sqlite database without errors? It's possible that the file itself wasn't generated correctly during your analysis steps.

christinaread commented 1 year ago

Hi, thank you for your response! I'm using the most updated version and the path is correct.

I'm not sure the sqlite file is merged without errors--is there a way to check this? Originally, I was using the .zip output file from the hospital in the function CohortDiagnostics::CreateMergedResultsFile, but this generated an database of empty tables every time. Now I am using an sqlite file generated by the hospital that is not empty, but I'm not sure that it is in the correct format.

azimov commented 1 year ago

Hi Christina, there should be a log file of the execution stored in the output folder (the exportFolder parameter to executeDiagnostics) the default name is just log.txt. Uploading this as a gist and attaching the link may help me see if anything went wrong.

christinaread commented 1 year ago

I am running the function CohortDiagnostics::launchDiagnosticsExplorer using either the zip files or the sqlite files (we tried both but neither worked) provided by the hospital, who ran the executeDiagnostics function. Is there a way for me to access a log.txt of the error when I am not the one also running executeDiagnostics?

azimov commented 1 year ago

@christinaread as it was someone external that ran this study, I think that you will need to contact the site to get the log file directly. There shouldn't be any patient sensitive information contained within but we can't verify this in the same way as we do for the csv files.

If you have the zip file looking at the CSV files there may be beneficial. metadata.csv and executionTimes.csv will give indications as to what was executed, what the parameters were and how long it took. This information shouldn't contain any sensitive information so you may be able to share it.

azimov commented 1 year ago

closed due to inactivity