Predicting Generalization in Deep Learning Forum

Go back to competition Back to thread list Post in this thread

> float division by zero in Score.py

Hi
I have the following error on phase 2:

WARNING: Your kernel does not support swap limit capabilities or the cgroup is not mounted. Memory limited without swap.
Traceback (most recent call last):
File "/tmp/codalab/tmp401a0fq1/run/program/score.py", line 218, in
task_average_score = sum(task_scores)/float(len(task_scores))
ZeroDivisionError: float division by zero

My code runs fine on phase 1. What can I do ?
The deadline is approaching and I was not able to submit something that do not fail (after 4 submissions, including 3 failing due to lack of space on device). It is the same code as the Phase 1 (and it never failed before !).

The November 1st, will it be possible to send a list of solutions to organizers (.zip or timestamp of submitted solutions) to ensure that they will be evaluated ? I submitted more than 3 times already :(

Posted by: AlgueRythme @ Oct. 28, 2020, 2:28 p.m.

Your submission ran out of time on the private data. In general you can keep submitting if fewer than 3 submissions have completed.
In this case, we will run them on increased time limit if it didn't time out on development phase but timed out on the test phase.
In any case, you would not need to send us the solutions.

Posted by: ydjiang @ Oct. 28, 2020, 3:15 p.m.
Post in this thread