Do all the things like ++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatarSign Up
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple APILearn More
McSebi514182dDecimal is built on top of floats, just work with numpy
Dont even know what you want to achieve
Wisecrack7378182d@McSebi if decimal is built on floats, how is it that se can set precision higher in Decimal than standard floats support?
Reason I'm asking is because I may have to handle variables with more than 128bit precision during training and classification of some data.
Lucky-Loek437182dWhy would you need such insane precision in machine learning? If 64 bit float isn't enough, it seems to me you're solving the wrong problem
Wisecrack7378179d@Lucky-Loek need greater than 64 bit precision, because the training data has entries with significantly more digits than 64 bits will cover.
There's no half measure here. Although I considered conversion to strings of characters, or chunking, or some sort of hashing/minifying scheme. I wanted to try the naive approach first instead of prematurely optimizing.
electrineer27400179dSounds like there is one you make one
geneticCode1398167dMy next favourite nerd. When you find it let us know.
Anyone know of a machine learning library for python, that supports Decimal module?