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
{{ message }}
Repository navigation
Participant %7: Team CSV, Universidad Central "Marta Abreu" de Las Villas #20
Thanks @monperrus!!
However, my approach has some performance issues. For instance, it takes almost 2h for Dataset1, which is far from the performance results of @tdurieux. Also, I think the accuracy (in terms of the loss function) should be improved much more to really win the competition. I'll continue working on that.
Strangely my technique is still better for the dataset 2 but worse for the others.
I still have some room for improvement but I am very happy of the performance of my technique. It takes less than 10min to have the results on all datasets. That is helping a lot to try new improvements
Activity
monperrus commentedon May 18, 2018
cesarsotovalero commentedon May 18, 2018
My current scores using just a very naive string comparison based approach:
Score on dataset1: 0.1236735
Score on dataset2: 0.1096176
No machine learning yet.
monperrus commentedon May 21, 2018
Yes. The first 0.8 are easy to get (purely due to the data).
The remaining points are super hard.
Best score seen so far:
cesarsotovalero commentedon May 29, 2018
My last scores:
My current approach: string matching + parse checking
A related paper: A comparison of code similarity analysers
chenzimin commentedon May 30, 2018
Thanks, I have updated the rankings
monperrus commentedon May 31, 2018
cesarsotovalero commentedon Aug 21, 2018
Hi everyone, I want to give an update of my scores for the preliminary ranking:
My current approach is: string matching + parse checking + decision rules + heuristics
monperrus commentedon Aug 22, 2018
cesarsotovalero commentedon Aug 22, 2018
Thanks @monperrus!!
However, my approach has some performance issues. For instance, it takes almost 2h for Dataset1, which is far from the performance results of @tdurieux. Also, I think the accuracy (in terms of the loss function) should be improved much more to really win the competition. I'll continue working on that.
tdurieux commentedon Aug 22, 2018
Strangely my technique is still better for the dataset 2 but worse for the others.
I still have some room for improvement but I am very happy of the performance of my technique. It takes less than 10min to have the results on all datasets. That is helping a lot to try new improvements