i have a dataframe(df) of 100000 rows. i want to convert to a netcdf. i know the best way is xarray, but my df is bit complicated as follows
`*a b c d e f*
[1,2,3...n] 4 5 8 [2,3...m] [m*n matrix]
[1,2,3...n] 5 5 8 [2,3...m] [m*n matrix]
[1,2,3...n] 4 5 10 [2,3...m] [m*n matrix]
.....100 k rows`
in the above a and e colomn are always the same for all rows.** b,c,d** colomns are repeatation of the values. now i want to create a netcdf where the attribute f is a 5-dimensional multidimension array of size mn*(uniques_values of b)*(uniques_values of c) (uniques_values of d)
since the number of rows are too many. it is better to avoid looping and use concurrancey
any good way is helpful, your text thanks in advance
i tried using the following but error
ds = df.set_index(['b', 'c', 'd']).to_xarray()
ds.to_netcdf('~/test.nc')