You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
the format of positive and negative sample files is the B/W values (can be floats from 0 to 1) of each sample, one per line. E.g. If base resolution is 21x21, each line of the samples file should have 441 float values:
Should the images be of size 21x21? Because when I try to load the training images, I get the following error followed by a segmentation fault:
"Warning: Image size differs from 21x21 (ignoring sample)"
Also, what should be the format of the POS_SAMPLE_FILE and NEG_SAMPLE_FILE be?
I created 2 text files as described below:
POS_SAMPLE_FILE.txt
image_path number_of_object_instances object_location_in_the_image
NEG_SAMPLE_FILE.txt
image path
The text was updated successfully, but these errors were encountered: