Closed alexdesiqueira closed 4 years ago
For the record, code works properly when not using -p
/-pp
:
$ python pipeline.py -i butterfly/tests/test_files/test_input/BMNHE_500607.JPG
Image 1/1 : BMNHE_500607.JPG
Lossy conversion from float32 to uint8. Range [0, 1]. Convert image to uint8 prior to saving to suppress this warning.
Lossy conversion from float32 to uint8. Range [0, 1]. Convert image to uint8 prior to saving to suppress this warning.
left_wing : 13.61 mm
right_wing : 13.66 mm
left_wing_center : 17.31 mm
right_wing_center : 17.28 mm
wing_span : 28.98 mm
Yes, I had the same issue.
This should be fixed now; please verify.
Thanks, Stéfan. I'm having issues with repeated executions, though; an example.
$ python pipeline.py -s 'measurements' -pp -i butterfly/tests/test_files/test_input/BMNHE_500607.JPG
Image 1/1 : BMNHE_500607.JPG
left_wing : 13.57 mm
right_wing : 13.66 mm
left_wing_center : 17.3 mm
right_wing_center : 17.29 mm
wing_span : 28.98 mm
$ python pipeline.py -s 'measurements' -pp -i butterfly/tests/test_files/test_input/BMNHE_500607.JPG
Image 1/1 : BMNHE_500607.JPG
left_wing : 13.57 mm
right_wing : 13.66 mm
left_wing_center : 17.3 mm
right_wing_center : 17.29 mm
wing_span : 28.98 mm
I have to remove cachedir
and output
after each run to generate new results. Let me test with more than one image.
Yeap, check the commit log :)
Saw it now. Sorry about that :sweat_smile: Closing this then. Solved by fe0cb8. Thank you Stéfan!
I can't generate a plot when combining
-p
/-pp
and-s 'measurements'
. Code breaks when runningpipeline.py
, line 197:points_interest = tracing.main(bfly_bin, axes)
. it raises this issue when receivingaxes=None
.Command used:
$ python pipeline.py -pp -i butterfly/tests/test_files/test_input/BMNHE_500607.JPG
(part of the) Traceback:
Maybe there's something wrong with
joblib
...? Tested with twojoblib
versions: 0.14.1, 0.15.1. Thank you!