earnestt1234 / fed3_viz Goto Github PK

Taken from the FEDForum:

Hi all,

I found the FED3 Viz application to analyze the data collected from the FED3 device. First of all, I’d like to say that using this application has simplified the analysis to a great extent and that it has been extremely convenient.

However, I noticed that when I want to get information about the day- and nighttime data (pellets, pokes, etc.), it always analyzes the data according to a fixed daytime from 7 AM to 7 PM and nighttime from 7 PM to 7 AM. I changed the settings “Shade dark periods (lights on/off)” to 8 AM and 8 PM (according to our local facility). But I think this only applies to the plots and not the Summary Stats? Is there a way to change the day- and nighttime for the whole application?

Any suggestions would be appreciated.

Thank you,

Chrissi

Relevant code section is here:

Hi Tom! Thank you for this amazing work. I am currently trying to load in some of my FED data to plot. I found that both your example data from the github and my own data only allow me to plot some of the option (i.e. single poke plot is fine, but average poke plot is not).

Any ideas why this might be?

Hi. I cannot load a .csv or .xlx file despite the fact that all the other files in this group are loading. The error message reads

ValueError: NaTType does not support strftime

Additionally, the option to create line chronograms in not populating. I have worked with this dataset successfully before. But, after re-extracting fed3 viz .exe for the sixth time (separate issue) the line chronogram will not work.

Help with this issue is much appreciated.

The program crashes when making average plots for groups which have different amount of bins when using the "shared date & time averaging". This is due to the data retrieval function in getdata.py: there is a filtering step which was included at the incorrect indentation, preventing its application. In plots.py, the indentation is correct and the problem doesn't occur. So the plot can be created, but when creating the associated DataFrame of data, there is a problem.

This is baked into the program, and is unintended behavior (the non-overlapping dates should just be ignored. Some workarounds are as follows:

1. Use the date filter functionality to manually set the range of data being plotted. This is basically doing the date windowing that the plot functions are supposed to do. Just be aware that this setting being on affects all plots created (so turn it off for other things if you don't want it on).
2. Use a different averaging scheme, like Settings > Averaging > Alignment method for averaging > shared time.
3. Manually go into the CSV files and delete the data that occur in non-overlapping windows.

The calculation of interpellet intervals appears to be incorrect, and will mark the first interpellet interval (in the row of the second pellet) as NaN. Later intervals appear to be correct. This amounts to one interpellet interval being lost, and also affects meal sizes.