issue_comments

### UPDATE 1: ### I have also tried to chunk over the nFlowElem dimension instead of 'time'. Then I define the input and output core dimensions as written below:

input_core_dims=[['time','laydim'],[],[],['dim_0','dim_1'],['dim_0','dim_1']], output_core_dims=[['dim_0','dim_1','time','laydim']], Now I "only" get an error concerning the size of my output (which is nolonger chunked). My output is nolonger chunked, because nFlowElem is lost after the interpolation. Consequently, the (600,560,1009,20) dataarray is too large for my work memory.

Is there perhaps a way to have apply_ufunc chunk your output along any of the other dimensions?

### UPDATE 2: ###

Alternatively, I have tried to chunk over the time dimension (50 time steps) and I have removed all input/output core dimensions.

And if I then define interp_to_grid as follows (to get the right input dimensions):

def interp_to_grid(u,xc,yc,xint,yint):

    xc = xc[:,0,0,0,0]
    yc = yc[:,0,0,0,0]
    u = u[:,:,:,0,0]
    print(u.shape,xc.shape,xint.shape) #input shape

    ug = griddata((xc,yc),u,(xint,yint), method='nearest', fill_value=np.nan)
    print(ug.shape) #output shape

    return ug

I do get the right dimensions for my interp_to_grid-input (see first line) and output (see second line). However, I get the ValueError: axes don't match array:

(194988, 50, 20) (194988,) (600, 560)
(600, 560, 50, 20)
Traceback (most recent call last):

  File "<ipython-input-11-6b65fe3dba5b>", line 1, in <module>
    uxg.loc[dict(time=np.datetime64('2014-09-18'),laydim=19)].plot()

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\xarray\plot\plot.py", line 444, in __call__
    return plot(self._da, **kwargs)

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\xarray\plot\plot.py", line 160, in plot
    darray = darray.squeeze().compute()

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\xarray\core\dataarray.py", line 899, in compute
    return new.load(**kwargs)

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\xarray\core\dataarray.py", line 873, in load
    ds = self._to_temp_dataset().load(**kwargs)

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\xarray\core\dataset.py", line 798, in load
    evaluated_data = da.compute(*lazy_data.values(), **kwargs)

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\base.py", line 565, in compute
    results = schedule(dsk, keys, **kwargs)

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\threaded.py", line 84, in get
    **kwargs

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\local.py", line 487, in get_async
    raise_exception(exc, tb)

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\local.py", line 317, in reraise
    raise exc

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\local.py", line 222, in execute_task
    result = _execute_task(task, data)

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\core.py", line 121, in _execute_task
    return func(*(_execute_task(a, cache) for a in args))

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\core.py", line 121, in <genexpr>
    return func(*(_execute_task(a, cache) for a in args))

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\core.py", line 121, in _execute_task
    return func(*(_execute_task(a, cache) for a in args))

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\optimization.py", line 963, in __call__
    return core.get(self.dsk, self.outkey, dict(zip(self.inkeys, args)))

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\core.py", line 151, in get
    result = _execute_task(task, cache)

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\core.py", line 121, in _execute_task
    return func(*(_execute_task(a, cache) for a in args))

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\dask\utils.py", line 35, in apply
    return func(*args, **kwargs)

  File "<__array_function__ internals>", line 6, in transpose

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\numpy\core\fromnumeric.py", line 658, in transpose
    return _wrapfunc(a, 'transpose', axes)

  File "C:\Users\920507\Anaconda3\envs\dfm_tools_env\lib\site-packages\numpy\core\fromnumeric.py", line 58, in _wrapfunc
    return bound(*args, **kwds)

ValueError: axes don't match array

@mathause Thank you for your response! I believe that map_blocks only works for functions that consume and return xarray objects, right? Scipy.interpolate.griddata unfortunately does not return Xarray object.

However, if there are other (xarray) functions which I could use to interpolate my results to a regular grid - I'd be very interested.