python - Storing Numpy or Pandas Data in Mongodb -


i'm trying decide best way store time series data in mongodb. outside of mongo i'm working them numpy arrays or pandas dataframes. have seen number of people (such in post) recommend pickling , storing binary, under impression pickle should never used long term storage. true data structures might have underlying code changes class structures? put way, numpy arrays stable fine pickle, pandas dataframes might go bad pandas still evolving?

update:

a friend pointed me this, seems start on want:

http://docs.scipy.org/doc/numpy/reference/routines.io.html

numpy has own binary file format, should long term storage stable. once working i'll come , post code. if else has made work i'll happily accept answer.

we've built open source library storing numeric data (pandas, numpy, etc.) in mongodb:

https://github.com/manahl/arctic

best of all, it's easy use, pretty fast , supports data versioning, multiple data libraries , more.


Comments

Popular posts from this blog

apache - Remove .php and add trailing slash in url using htaccess not loading css -

javascript - jQuery show full size image on click -