excise some old code from the RPI model that was making it fail, re: the old way of cropping out a detector slice

This commit is contained in:
levita_a committed 2024-06-26 11:09:43 +02:00
1 parent 29d14c9537
commit ee0d483f26
4 files changed
+17 -36

No files matched your search

-7
View File
@@ -236,7 +236,6 @@ class Bragg2DPtycho(CDIModel):
propagation_distance=None,
scattering_mode=None,
oversampling=1,
auto_center=False,
probe_fourier_crop=None,
propagate_probe=True,
correct_tilt=True,
@@ -255,12 +254,6 @@ class Bragg2DPtycho(CDIModel):
(indices, translations), patterns = dataset[:]
dataset.get_as(*get_as_args[0],**get_as_args[1])
# Set to none to avoid issues with things outside the detector
if auto_center:
center = tools.image_processing.centroid(t.sum(patterns,dim=0))
else:
center = None
# Then, generate the exit wave geometry from the dataset
ewg = tools.initializers.exit_wave_geometry
ew_basis = ewg(det_basis,
+11 -25
View File
@@ -60,7 +60,7 @@ class RPI(CDIModel):
def __init__(self, wavelength, detector_geometry, probe_basis,
probe, obj_guess, detector_slice=None,
probe, obj_guess,
background=None, mask=None, saturation=None,
obj_support=None, oversampling=1, weight_matrix=False,
simulate_finite_pixels=False):
@@ -83,7 +83,6 @@ class RPI(CDIModel):
scale_factor = t.tensor([probe.shape[-1]/obj_guess.shape[-1],
probe.shape[-2]/obj_guess.shape[-2]])
self.obj_basis = self.probe_basis * scale_factor
self.detector_slice = detector_slice
# Maybe something to include in a bit
# self.surface_normal = t.tensor(surface_normal)
@@ -117,13 +116,8 @@ class RPI(CDIModel):
# self.obj = t.nn.Parameter(obj_guess.to(t.float32))
if background is None:
if detector_slice is not None:
background = 1e-6 * t.ones(
self.probe[0][self.detector_slice].shape,
dtype=t.float32)
else:
background = 1e-6 * t.ones(self.probe[0].shape,
dtype=t.float32)
background = 1e-6 * t.ones(self.probe[0].shape,
dtype=t.float32)
self.background = t.tensor(background, dtype=t.float32)
@@ -139,7 +133,7 @@ class RPI(CDIModel):
@classmethod
def from_dataset(cls, dataset, probe, obj_size=None, background=None, mask=None, padding=0, n_modes=1, saturation=None, scattering_mode=None, oversampling=1, auto_center=False, initialization='random', weight_matrix=False, probe_threshold=0, simulate_finite_pixels=False):
def from_dataset(cls, dataset, probe, obj_size=None, background=None, mask=None, n_modes=1, saturation=None, scattering_mode=None, oversampling=1, initialization='random', weight_matrix=False, probe_threshold=0, simulate_finite_pixels=False):
wavelength = dataset.wavelength
det_basis = dataset.detector_geometry['basis']
@@ -153,21 +147,14 @@ class RPI(CDIModel):
_, patterns = dataset[:]
dataset.get_as(*get_as_args[0],**get_as_args[1])
# Set to none to avoid issues with things outside the detector
if auto_center:
center = tools.image_processing.centroid(t.sum(patterns,dim=0))
else:
center = None
# Then, generate the probe geometry from the dataset
ewg = tools.initializers.exit_wave_geometry
probe_basis, probe_shape, det_slice = ewg(det_basis,
det_shape,
wavelength,
distance,
center=center,
padding=padding,
oversampling=oversampling)
ew_basis = ewg(det_basis,
det_shape,
wavelength,
distance,
oversampling=oversampling)
if not isinstance(probe,t.Tensor):
probe = t.as_tensor(probe)
@@ -248,8 +235,8 @@ class RPI(CDIModel):
obj_support = probe_lr > t.max(probe_lr) * probe_threshold
obj_support = t.as_tensor(binary_dilation(obj_support))
return cls(wavelength, det_geo, probe_basis,
probe, obj_guess, detector_slice=det_slice,
return cls(wavelength, det_geo, ew_basis,
probe, obj_guess,
background=background, mask=mask, saturation=saturation,
obj_support=obj_support, oversampling=oversampling,
weight_matrix=weight_matrix,
@@ -359,7 +346,6 @@ class RPI(CDIModel):
# a 4D wavefield array as well as a 5D array.
m = tools.measurements.quadratic_background(wavefields,
self.background,
detector_slice=self.detector_slice,
measurement=tools.measurements.incoherent_sum,
saturation=self.saturation,
oversampling=self.oversampling,
+4 -2
View File
@@ -612,7 +612,7 @@ def calc_frc(im1, im2, basis, im_slice=None, nbins=None, snr=1., limit='side'):
#n_pix = n_pix / 4 # This is for an apodized image, apodized with a hann window
frc = np.abs(numerator) / np.sqrt(denominator_F1*denominator_F2)
frc = numerator / np.sqrt(denominator_F1*denominator_F2)
# This moves from combined-image SNR to single-image SNR
snr /= 2
@@ -1210,8 +1210,10 @@ def standardize_reconstruction_set(
ip.hann_window(obj_2[window]),
full['obj_basis'], nbins=nbins, limit=frc_limit)
# The correct formulation when the final output is the full reconstruction
ssnr = 2 * frc / (1 - frc)
ssnr = 2 * np.abs(frc) / (1 - np.abs(frc))
results = {
'obj_half_1': obj_1,
+2 -2
View File
@@ -147,12 +147,12 @@ def generate_high_NA_k_intensity_map(sample_basis, det_basis,det_shape,distance,
det_shape = np.array(tuple(det_shape))
try:
distance = distance.detach().cpu().numpy()[0]
distance = distance.detach().cpu().numpy().ravel()[0]
except:
pass
try:
wavelength = wavelength.detach().cpu().numpy()[0]
wavelength = wavelength.detach().cpu().numpy().ravel()[0]
except:
pass