Bpr fix out of bound error #13
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Within the BPR optimization loop, the pointer into the data can run out of bounds, because the condition in while loop only checks if one pointer of the examples (negative or positive) is still within reasonable bounds, but then the other pointer might be increased in the end. THis fix double checks that the pointer stays within bounds.
|
@felixmaximilian
Can you post your one pair breaking example here? I didn't get it from the debug output. I probability have time tomorrow to have a closer look in order to check if there are related issues. |
|
You can see the pair in the gdb info locals output in the issue report. It's called sample_row_p and n. But without the data it won't help you understand I think. Unfortunately I cannot post the feature matrix. |
|
@felixmaximilian |
|
@felixmaximilian |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Potential fix for ibayer/fastFM#61 (#61).
The problem with this while loop is, the following:
While p_n is the reason why the conditional OR evaluates to true (even the p_p part is false), i_to_update can decide to update p_p which leads to an infinite loop and the p_p accesses Ax[p_p] on an undefined memory slot.