Data preparation: Each row of the data set should be a record (i.e. an observation). Each column should be a variable respectively a parameter of interest. Do not store multiple variables in one column. Use distinct names for the columns. Missing values are given by empty cells. Other character values (e.g. "NA") are not recognized as missing values.
Once you have your data prepared in the specified way, save it with the CSV format. If your data file is in an Excel or LibreOffice format, you can try to convert it to CSV by selecting "Save As". Try to select the format CSV with the character encoding UTF-8, this helps reading the data correctly. You can also prepare or inspect the CSV file with a text editor like Notepad++.
When your data is ready, you can proceed with the upload. A couple of upload properties must be specified. If you are not sure about them, you can choose the option "Auto" for each property. In this case, the Statsomat app will try to identify automatically the right settings. A severe continuity condition will lead to more categorical variables than a relaxed continuity condition.
After a successful upload of the data, you can proceed with the selection of the variables for the EDA. Now you can generate and finally download the report.
Instructions to run the code: If not already available, install Python and the required packages. We recommend the Anaconda distribution and Spyder as the IDE. Copy the code, paste it in a new Python file. Change the path to your data in the line filepath = .... Run the code. Explore the results. Disclaimer: Due to differences between your local system and the app running on Shinyapps.io, the results generated locally by executing the delivered code may be different to the PDF report.