RuntimeError during auto-classification

[shbhuk]

Hi,
I get the following error while running the auto-classification in command line or in the GUI

It runs fine as long as I skip directly to the eye-checking, but otherwise gives me the error.

RuntimeError: wrapped C/C++ object of type FigureCanvasQTAgg has been deleted

Running the example file -
`python pyhammer.py -f -c -i test_case/exampleInputFile.txt

The results file is set to PyHammerResults.csv. Do you want to change it? (y/n): n
The reject file is set to RejectSpectra.csv. Do you want to change it? (y/n): n
Do you want to skip straight to eye-checking? (y/n): n
Do you want to supply a S/N cut when auto-classifying? (y/n): n
Do you want to have the spectral lines used for classification written to a file? (y/n): y
`

I'm using the following versions -

Tkinter == 8.6
tqdm == 4.46
matplotlib == 3.5.2
numpy==1.22.3

[edwindas4]

There might be some issues with "zoom" command mainly due to the updated version of matplotlib.

Try this: https://stackoverflow.com/questions/66000179/attributeerror-navigationtoolbar2qt-object-has-no-attribute-active

Also, you can try reducing the number of input stars.