In your example data, each value is represented with two bytes: one for the
value, one for a comma or newline. Each Int64 value is 8 bytes. If all your
values are between 0 and 255, you could use UInt8 to represent them and cut
the size in half.
On Tue, Jul 21, 2015 at 1:16 PM, paul analyst <paul.anal...@mail.com> wrote:
> I have data in txt file, some milons like this:
> 0,0,2,0,0,0,0,0,0,0,0,0,0,0,0,0,0
> 0,0,0,0,0,0,0,2,0,0,0,2,0,0,0,0,1
> 0,0,0,0,1,0,0,0,1,0,0,0,0,0,0,0,1
>
> Coding win1250.
>
> size of dane.txt is 1.3 GB
>
> D=readcsv("dane.txt")
> k,l=size(D)
>
> using HDF5, JLD
> hfi=h5open("D.h5","w")
> close(hfi)
>
> fid = h5open("D.h5","r+")
> g = fid["/"]
> dset1 = d_create(g, "/D", datatype(Int64), dataspace(k,l))
> dset1[:,:]=D
> close(fid)
>
> After save to h5 file the file has 6.3 GB ? Why new file is 4 times biger?
> Paul
>
