I have a small project where the working files have been saved as pickles using pandas 0.20, as a result this creates a backward compatibility issue with an older version of pandas that we need to downgrade to support another library.
This task will involve:
1) creating a script that will get all files in the `./data` directory and make copies of them to HDF5 format (using a compression that is compatible with all versions).
2) replacing all parts of the codebase that uses `read_pickle` to read from the HDF5 counterpart.
3) replace all parts of the codebase that uses `to_pickle` to write to HDF5.
February 24, 2018
I am looking for a mix of experience and value