def load_csv(activation_file, verbose=False):
if activation_file is None:
print("WARNING: No activations loaded since file not specified; i.e. there is no ground truth.")
return
# read csv
with open(activation_file) as csv_file:
csv_reader = csv.reader(csv_file)
line_count = 0
id_frame_xyz_camval = []
for row in csv_reader:
if verbose and line_count == 0:
print(row)
elif line_count >= 1:
id_frame_xyz_camval.append(torch.tensor(
(float(row[0]), float(row[1]), float(row[2]), float(row[3]), float(row[4]), float(row[5]))))
line_count += 1
id_frame_xyz_camval = torch.stack(id_frame_xyz_camval, 0)
gt = em.EmitterSet(xyz=id_frame_xyz_camval[:, 2:5], frame_ix=id_frame_xyz_camval[:, 1].long(),
phot=id_frame_xyz_camval[:, -1], id=id_frame_xyz_camval[:, 0].long())
gt.sort_by_frame_()
return gt
def test_forward_handcrafted(self, evaluator):
# if evaluator.mode != 'phot':
# return
"""Setup"""
tp = em.EmitterSet(xyz=torch.zeros((4, 3)),
phot=torch.tensor([1050., 1950., 3050., 4050]),
frame_ix=torch.tensor([0, 0, 1, 2]),
bg=torch.ones((4, )) * 10,
xy_unit='px',
px_size=(127., 117.))
ref = tp.clone()
ref.xyz += 0.5
ref.phot = torch.tensor([1000., 2000., 3000., 4000.])
ref.xyz_cr = (torch.tensor([[10., 10., 15], [8., 8., 10], [6., 6., 7],
[4., 4., 5.]]) / 100.)**2
ref.phot_cr = torch.tensor([10., 12., 14., 16.])**2
ref.bg_cr = torch.tensor([1., 2., 3., 4])**2
"""Run"""
_, _, _, dpos, dphot, dbg = evaluator.forward(
tp, ref) # test only on non reduced values
"""Assertions"""
assert (dpos.abs().argsort(0) == torch.arange(4).unsqueeze(1).repeat(1, 3)).all(), "Weighted error for pos." \
"should be monot. increasing"
assert (dphot.abs().argsort(descending=True) == torch.arange(4)).all(), "Weighted error for photon should be " \
"monot. decreasing"
assert (dbg.abs().argsort(descending=True) == torch.arange(4)).all(), "Weighted error for background should be " \
"monot. decreasing"
def em(self):
"""Setup"""
xyz = torch.rand(100, 3) * torch.Tensor([[100., 100., 1000.]])
return emitter.EmitterSet(xyz,
xyz_sig=xyz * 0.1,
phot=torch.ones(100),
frame_ix=torch.arange(100),
xy_unit='nm')
def test_save_load_h5py(em_rand, em_all_attrs, save_fn, load_fn, extension,
tmpdir):
path = str(tmpdir / f'emitter{extension}')
for em in (em_rand, em_all_attrs):
save_fn(path, em.data, em.meta)
data, meta, decode_meta = load_fn(path)
em_reloaded = emitter.EmitterSet(**data, **meta)
assert em == em_reloaded # if equality check is wrong, this is wrong as well
assert decode_meta['version'][0] == 'v'
def test_weight_hard(self, waiter):
"""
Tests entire weight unit with hard computed values
Args:
waiter: fixture
"""
"""Setup"""
tar_frames = torch.zeros((1, 6, 5, 5))
tar_frames[:, 5] = torch.rand_like(tar_frames[:, 5]) # let bg be non-zero
em = emitter.EmitterSet(xyz=torch.tensor([[1., 1., 0], [3., 3., 0.]]), phot=torch.Tensor([1., 5.]),
frame_ix=torch.tensor([0, 0]), xy_unit='px')
"""Run"""
mask = waiter.forward(em, tar_frames, 0, 0)
"""Assertions"""
assert (mask[:, 0] == 1.).all(), "p channel must be weight 1"
# some zero value assertions applicaple for both const and phot weight
assert (mask[:, 1:-1, 2, 2] == 0.).all(), "intersection must be 0"
assert (mask[:, 1:-1, 3:, :2] == 0.).all()
assert (mask[:, 1:-1, :2, 3:] == 0.).all()
if waiter.weight_mode == 'const':
assert (mask[:, 5] == 1.).all(), "bg channel must be weight 1"
assert (mask[:, 1:-1].unique() == torch.tensor([0., 1])).all(), "in const. mode values must be 0 or 1"
if waiter.weight_mode == 'phot':
assert (mask[:, 1:-1, :2, :2] == 1.).all(), "CRLB estimate for photon count 1"
assert mask[:, 1, 3, 3] == pytest.approx(0.02468, abs=0.0001), "Photon CRLB estimate for count of 5"
assert mask[:, 2, 3, 3] == pytest.approx(40.51641, abs=0.0001), "X CRLB estimate"
assert mask[:, 3, 3, 3] == pytest.approx(40.51641, abs=0.0001), "Y CRLB estimate"
assert mask[:, 4, 3, 3] == pytest.approx(40.51641, abs=0.0001), "Y CRLB estimate"
assert test_utils.tens_almeq(mask[:, 5], 1 / tar_frames[:, 5] ** 2.3, 1e-5), "BG CRLB estimate"