Get field on CPU explicitly for show/write_to_file

lasy/laser.py

@@ -1,4 +1,6 @@
+import numpy as np
 from .backend import xp
+
 from axiprop.lib import PropagatorFFT2, PropagatorResampling
 from scipy.constants import c
 
@@ -342,11 +344,12 @@ def show(self, **kw):
         ----------
         **kw: additional arguments to be passed to matplotlib's imshow command
         """
-        temporal_field = self.grid.get_temporal_field()
+        # Get field on CPU
+        temporal_field = self.grid.get_temporal_field(to_cpu=True)
         if self.dim == "rt":
             # Show field in the plane y=0, above and below axis, with proper sign for each mode
             E = [
-                xp.concatenate(
+                np.concatenate(
                     ((-1.0) ** m * temporal_field[m, ::-1], temporal_field[m])
                 )
                 for m in self.grid.azimuthal_modes

lasy/utils/openpmd_output.py

@@ -53,7 +53,8 @@ def write_to_openpmd_file(
         Whether the envelope is converted to normalized vector potential
         before writing to file.
     """
-    array = grid.get_temporal_field()
+    # Get field on CPU
+    array = grid.get_temporal_field(to_cpu=True)
 
     # Create file
     series = io.Series("{}_%05T.{}".format(file_prefix, file_format), io.Access.create)
@@ -76,7 +77,7 @@ def write_to_openpmd_file(
         m.axis_labels = ["t", "r"]
 
     # Store metadata needed to reconstruct the field
-    m.set_attribute("angularFrequency", 2 * xp.pi * c / wavelength)
+    m.set_attribute("angularFrequency", 2 * np.pi * c / wavelength)
     m.set_attribute("polarization", pol)
     if save_as_vector_potential:
         m.set_attribute("envelopeField", "normalized_vector_potential")
@@ -91,20 +92,20 @@ def write_to_openpmd_file(
         }
 
     if save_as_vector_potential:
-        array = field_to_vector_potential(grid, 2 * xp.pi * c / wavelength)
+        array = field_to_vector_potential(grid, 2 * np.pi * c / wavelength)
 
     # Pick the correct field
     if dim == "xyt":
         # Switch from x,y,t (internal to lasy) to t,y,x (in openPMD file)
         # This is because many PIC codes expect x to be the fastest index
-        data = xp.transpose(array).copy()
+        data = np.transpose(array).copy()
     elif dim == "rt":
         # The representation of modes in openPMD
         # (see https://github.com/openPMD/openPMD-standard/blob/latest/STANDARD.md#required-attributes-for-each-mesh-record)
         # is different than the representation of modes internal to lasy.
         # Thus, there is a non-trivial conversion here
         ncomp = 2 * grid.n_azimuthal_modes - 1
-        data = xp.zeros((ncomp, grid.npoints[0], grid.npoints[1]), dtype=array.dtype)
+        data = np.zeros((ncomp, grid.npoints[0], grid.npoints[1]), dtype=array.dtype)
         data[0, :, :] = array[0, :, :]
         for mode in range(1, grid.n_azimuthal_modes):
             # cos(m*theta) part of the mode
@@ -113,12 +114,12 @@ def write_to_openpmd_file(
             data[2 * mode, :, :] = -1.0j * array[mode, :, :] + 1.0j * array[-mode, :, :]
         # Switch from m,r,t (internal to lasy) to m,t,r (in openPMD file)
         # This is because many PIC codes expect r to be the fastest index
-        data = xp.transpose(data, axes=[0, 2, 1]).copy()
+        data = np.transpose(data, axes=[0, 2, 1]).copy()
 
     # Define the dataset
     dataset = io.Dataset(data.dtype, data.shape)
     env = m[io.Mesh_Record_Component.SCALAR]
-    env.position = xp.zeros(len(dim), dtype=xp.float64)
+    env.position = np.zeros(len(dim), dtype=np.float64)
     env.reset_dataset(dataset)
     env.store_chunk(data)