Viewing 7 posts - 1 through 7 (of 7 total)
• Author
Posts
• #328

Hi moderator,

According to the provided penalty score calculation formula, the results of our calculation are different from the spreadsheet . Can you confirm it and provide a detailed calculation formula in the spreadsheet?

Thanks.

#332
kjd02002
Moderator

Dear ChoochooTrain,

Please provide an example of the discrepancy and we will investigate.

Thank you.

#333

Hi moderator,

Thanks for your quick reply.
Take T1 as an example.Except for the last cycle, all cycles are correct.We set the crack length of last cycle number to 7, which is actually 7.46. According to the spreadsheet calculation, penalty score is about 4.333.But according to the formula provided, T = 2+10*7.46 = 76.6, A= exp (2.3) +1 ≈ 9.974, M = 1.So the penalty score we calculated is S = T*A*M ≈764.

Thanks.

#336
kjd02002
Moderator

Hi ChoochooTrain,

Thank you for the example. I believe the discrepancy you’ve identified is that the penalty score is based on normalized crack length (% of final crack length) rather than absolute crack length (millimeters). For your example, this will result as follows: T=2+10*(7.46/7.46)=12, A=exp(|7/7.46-7.46/7.46|/0.2)-1=0.3611, M=1 –> S=T*A*M=4.333. This is not clearly identified in the penalty score description so we will update accordingly. Please let me know if this resolves your question.

Please note: normalized crack length is used in the penalty score calculation to ensure predictions for each sample has similar weighting regardless of final crack length. Teams will submit predicted crack length in millimeters (mm) and the penalty score will automatically be calculated based on the normalized crack length (% of final crack length).

Thank you.

• This reply was modified 1 year, 9 months ago by samuel_admin.
#488
realai
Member

Hi moderator,
If the true length is 0, but our estimation is not 0, what will the result be?

Thanks.

#489
realai
Member

Hi moderator,
I have another question: the value of time penalty function is always 12?

Thanks.

#492
realai
Member

Hi moderator,
We understood wrong before, we have got it.
Thanks!

Viewing 7 posts - 1 through 7 (of 7 total)
• The forum ‘2019 Data Challenge: Technical Questions’ is closed to new topics and replies.