problem in get_training_sets

[QNeuron]

Hi again,

I am trying to train the classifier with that command :

(base) C:\Users\dpag0680>python D:\Work\Code\gitDev\twoptb\twoptb\ROIs\train_roi _finder.py -rad 7 -shifts 3 7 ferretclassifier1 D:\Work\Data\twopi\rawFerret\Nal a\processed\Nala_area02

and got the following error:

C:\Users\dpag0680\AppData\Local\Continuum\anaconda2\lib\site-packages\h5py__ini t__.py:36: FutureWarning: Conversion of the second argument of issubdtype from float to np.floating is deprecated. In future, it will be treated as np.floa t64 == np.dtype(float).type. from ._conv import register_converters as _register_converters D:\Work\Data\twopi\rawFerret\Nala\processed\Nala_area02\ROIs\area02-registered_d ata-(20180520_02_17_51)-_area02_tones_ROIs.p D:\Work\Data\twopi\rawFerret\Nala\p rocessed\Nala_area02\Nala_area02.h5

C:\Users\dpag0680\AppData\Local\Continuum\anaconda2\lib\site-packages\skimage\ut il\dtype.py:122: UserWarning: Possible precision loss when converting from float 64 to uint16 .format(dtypeobj_in, dtypeobj_out)) Traceback (most recent call last): File "D:\Work\Code\gitDev\twoptb\twoptb\ROIs\train_roi_finder.py", line 223, i n boutons,non_boutons = get_training_sets(sets,rad,shifts=eg_shift) File "D:\Work\Code\gitDev\twoptb\twoptb\ROIs\train_roi_finder.py", line 141, i n get_trainingsets if np.max(t)==0: UnboundLocalError: local variable 't_' referenced before assignment

[yves-weissenberger]

should be fixed in recent push?

[QNeuron]

Yep, Actually the hdf files were corrupted (presumably by some of the multiple crashes of the functions ?). I re-generated them from the raw data, and the classifier works now (both training and applying). There's still an issue, but that's for another issue thread :)