Hi,
In data/scripts/kaist_to_voc_format.py, I see the following:
# all train images but 20% from test
if set_ == 'set06' and set_counter == 901: # all = 1161, person only = 901
But, in the annotations set06 folder, I see 5562 text files that have 'person'. 20% of 5562 comes to 1112. But the above logic is checking the counter for 901.
Hi, In data/scripts/kaist_to_voc_format.py, I see the following:
But, in the annotations set06 folder, I see 5562 text files that have 'person'. 20% of 5562 comes to 1112. But the above logic is checking the counter for 901.
Can someone clarify? Thank you.