- This topic has 2 replies, 2 voices, and was last updated 3 years, 2 months ago by .
-
Posts
-
Hi,
I am a student starting to study computer science, and decided to create a function that finds all the correct and incorrect digit sums within a specified range. When I ran the function for the first hundred numbers I found that there were far more incorrect digit sums than correct ones. Given that the accuracy of the digit sum is 90%, please explain this finding. The following file contains the function, DS.ipynb
Thank you,
Akiva Neuman
Hi Akiva,
The denominator in your code does not include cases where the DS method correctly predicts a calculation as incorrect.
I have updated your code to reflect this and you can find the notebook here: https://github.com/abhivr/public_repo/blob/main/Digit_Sum_Checker.ipynb
When I wrote the book & created the course, I estimated the accuracy of the DS method using simple probability.
The digit sum can only be a number between 1 to 9. So there are 9 possible values.
8 of these 9 possible DS values will help you catch an error in the calculation without an issue.
One out of these 9 possible values is the DS of the correct answer but it can also be the DS of the incorrect answer.
So there are actually a total of 10 possible values for the Digit sum (9 Correct DS + 1 Incorrect DS) with one number repeating in Correct DS and Incorrect DS.
So 9 / 10 (90%) the DS method correctly predicts the calculation as incorrect & correct. But 1/10 times the DS method predicts the calculation is correct when it is actually incorrect.
Hope this clarifies.
- You must be logged in to reply to this topic.