A core feature of the UniApp is the ability to plan out your analysis before you execute it. Firstly we plan out the analysis using placeholders and once we have a clear plan in place we proceed to perform the analysis. This ensures that the analysis is focused, organized and well executed. Additionally, a senior data scientist can plan out an analysis and then delagate the execution of the analysis to a junior data scientist.
TABLE OF CONTENTS
Below is an example of an analysis track with placeholder plots:
1. Creating a placeholder plot
Start by clicking on the create plot icon. This will lead you to the Create plot page.
You can start by filling in the plot name, plot description and selection the algorithm you plan to run. After this you can create a placeholder plot by clicking on the Save placeholder button.
After saving the placeholder you will be redirected back to the track page where you can see the placeholder you just save as a static plot. You can do this multiple times to plan out your analysis.
2. Running an algorithm from the placeholder plot
To run an algorithm from the placeholder plot click on the Details button of placeholder plot. From here you can edit the plot name and description or upload a new thumbnail. But most importantly from you can perform the analysis from the placeholder by clicking the Perform analysis button.
After clicking the Perform analysis button you be redirected to the Create plot page where the plot name, plot description and choice of algorithm is the same as it was in the placeholder plot.
From here you can select the data, set algortihm parameters and finally run the algorirthm by clicking the Run button. After running the algorithm the placeholder static plot will be replaced with an interactive plot type.