Open pixixiaonaogou opened 6 years ago
Do you know how to calculate recall and precision for a detection task?
there is a script in icdar's website
@argman I could not find it . Could you give me a url?
you need first register , when you submit result , you can find it
@argman it is a code or score?
@argman I can compute F1 score online, but script on the [webset] (http://rrc.cvc.uab.es/?ch=4&com=mymethods&task=1) cant successfully do it. Can you give some advice to figure it out? Thanks in advance.
@argman I can compute F1 score online, but script on the [webset] (http://rrc.cvc.uab.es/?ch=4&com=mymethods&task=1) cant successfully do it. Can you give some advice to figure it out? Thanks in advance.
Hi, have you solve the problem
@argman I can compute F1 score online, but script on the [webset] (http://rrc.cvc.uab.es/?ch=4&com=mymethods&task=1) cant successfully do it. Can you give some advice to figure it out? Thanks in advance.
The problem has been solved and I just pip2 install Polygon2 and use the command python2 script.py -g=gt.zip -s=1_submit.zip -o=./
I get this error when I run this command:python script.py –g=gt.zip –s=submit_nvlink_0.zip. I check the img_100.txt and it's ok, have someone met this problem or known what is the problem?
I get this error when I run this command:python script.py –g=gt.zip –s=submit_nvlink_0.zip. I check the img_100.txt and it's ok, have someone met this problem or known what is the problem?
In script, default_evaluation_params(), it expects a name in the form resimg#.txt instead of img#.txt
I get this error when I run this command:python script.py –g=gt.zip –s=submit_nvlink_0.zip. I check the img_100.txt and it's ok, have someone met this problem or known what is the problem?
In script, default_evaluation_params(), it expects a name in the form resimg#.txt instead of img#.txt
Thanks a lot, I have solved this problem.
how do find accuracy and how to get false positive,false negative, true positive & true negative scores for the images i pass. Please help me in doing this. Thnak you in advance
Hi. did you find out how to get them? thank you
Do you have the code the F1 score code to evaluate your model?