Explore >> Select a destination
|
You are here 
|
www.jamesridgway.co.uk | ||
| | | | |
backdrifting.net
|
|
| | | | | [AI summary] The author explains what Bloom Filters are, their limitations, various applications like pre-caching, and the mathematical theory behind using them as constant-space probabilistic data structures. | |
| | | | |
ketanvijayvargiya.com
|
|
| | | | | Bloom filters. Set membership. 1D array with k hash functions. False positives possible but no false negatives. Applications: One hit wonders can take up to 75% of cache space. BF can help identify so that you can skip caching such items and save cache space. Check for weak passwords, malicious URLs, username etc. If you want 100% accuracy, check in BF first and fallback to the real data store as required. In a NoSQL type database, check BF on whether an item exists before going to the disk. Therefore, decrease disk access. Count min sketch. Estimate frequency of all elements in a set. 2D array. Each row is for a given hash function, so R hash functions. These functions give a number between 0 and C-1, where C is the number of columns. Once all elements are ... | |
| | | | |
connorboyle.io
|
|
| | | | | TL;DR: if you are using scikit-learn 1.3.X and use f1_score() or classification_report() with the argument zero_division=1.0 or zero_division=np.nan1, then there's a chance that the output of that function is wrong (possibly by any amount up to 100%, depending on the number of classes in your dataset). E.g. for zero_division=1.0: In this post, np.nan refers to numpy.nan? | |
| | | | |
blog.lewman.com
|
|
| | | TL;DR I finally found a few hours to setup 'secure boot' on... | ||
