which is faster for load: pickle or hdf5 in python
I would consider only two storage formats: HDF5 (PyTables) and Feather
Here are results of my read and write comparison for the DF (shape: 4000000 x 6, size in memory 183.1 MB, size of uncompressed CSV - 492 MB).
Comparison for the following storage formats: (CSV, CSV.gzip, Pickle, HDF5 [various compression]):
read_s write_s size_ratio_to_CSVstorageCSV 17.900 69.00 1.000CSV.gzip 18.900 186.00 0.047Pickle 0.173 1.77 0.374HDF_fixed 0.196 2.03 0.435HDF_tab 0.230 2.60 0.437HDF_tab_zlib_c5 0.845 5.44 0.035HDF_tab_zlib_c9 0.860 5.95 0.035HDF_tab_bzip2_c5 2.500 36.50 0.011HDF_tab_bzip2_c9 2.500 36.50 0.011But it might be different for you, because all my data was of the datetime dtype, so it's always better to make such a comparison with your real data or at least with the similar data...