We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
added `code` formatting to --noconstant flag embedded in text
--invert_hash mentioned
Updated Feature Hashing and Extraction (markdown)
Adding a note about determining the number of features
'&' is a bitwise AND operation; the hashing code is: size_t word_hash = (p->hasher(feature_name,channel_hash)) & mask;
fix markdown of headings
reword, refactor, and less repetition
Added explanation to --hash option, updated murmur hash version and history.
Add the actual option for --readable_model
updated the bitmax info
how to get back feature weights!