def test_shepp_logan_SenseRecon_with_comm(self):
img, mps, ksp = self.shepp_logan_setup()
lamda = 0
comm = sp.Communicator()
ksp = ksp[comm.rank::comm.size]
mps = mps[comm.rank::comm.size]
img_rec = app.SenseRecon(
ksp, mps, lamda, comm=comm, alg_name='ConjugateGradient',
show_pbar=False).run()
npt.assert_allclose(img, img_rec, atol=1e-3, rtol=1e-3)
img_rec = app.SenseRecon(
ksp, mps, lamda, alg_name='GradientMethod',
show_pbar=False).run()
npt.assert_allclose(img, img_rec, atol=1e-3, rtol=1e-3)
img_rec = app.SenseRecon(
ksp,
mps,
lamda,
alg_name='PrimalDualHybridGradient',
max_iter=1000,
show_pbar=False).run()
npt.assert_allclose(img, img_rec, atol=1e-3, rtol=1e-3)
def jsens_calib(ksp, coord, dcf, ishape, device = sp.Device(-1)):
img_s = nft.nufft_adj([ksp],[coord],[dcf],device = device,ishape = ishape,id_channel =True)
ksp = sp.fft(input=np.asarray(img_s[0]),axes=(1,2,3))
mps = mr.app.JsenseRecon(ksp,
mps_ker_width=12,
ksp_calib_width=32,
lamda=0,
device=device,
comm=sp.Communicator(),
max_iter=10,
max_inner_iter=10).run()
return mps
def test_sense_model_with_comm(self):
img_shape = [16, 16]
mps_shape = [8, 16, 16]
comm = sp.Communicator()
img = sp.randn(img_shape, dtype=np.complex)
mps = sp.randn(mps_shape, dtype=np.complex)
comm.allreduce(img)
comm.allreduce(mps)
ksp = sp.fft(img * mps, axes=[-1, -2])
A = linop.Sense(mps[comm.rank::comm.size], comm=comm)
npt.assert_allclose(A.H(ksp[comm.rank::comm.size]), np.sum(
sp.ifft(ksp, axes=[-1, -2]) * mps.conjugate(), 0))
def test_shepp_logan_SenseRecon_with_comm(self):
img, mps, ksp = self.shepp_logan_setup()
lamda = 0
comm = sp.Communicator()
ksp = ksp[comm.rank::comm.size]
mps = mps[comm.rank::comm.size]
for solver in ['ConjugateGradient',
'GradientMethod',
'PrimalDualHybridGradient',
'ADMM']:
with self.subTest(solver=solver):
img_rec = app.SenseRecon(
ksp, mps, lamda, comm=comm, solver=solver,
show_pbar=False).run()
npt.assert_allclose(img, img_rec, atol=1e-2, rtol=1e-2)
parser.add_argument('ksp_file', type=str)
parser.add_argument('coord_file', type=str)
parser.add_argument('dcf_file', type=str)
parser.add_argument('mps_file', type=str)
args = parser.parse_args()
logging.basicConfig(level=logging.DEBUG)
logging.info('Reading data.')
ksp = np.load(args.ksp_file, mmap_mode='r')
coord = np.load(args.coord_file)
dcf = np.load(args.dcf_file)
# Choose device
comm = sp.Communicator()
if args.multi_gpu:
device = comm.rank
else:
device = args.device
logging.info('Jsense Recon.')
ksp = np.array_split(ksp, comm.size)[comm.rank]
mps = mr.app.JsenseRecon(ksp,
coord=coord,
weights=dcf,
mps_ker_width=args.mps_ker_width,
ksp_calib_width=args.ksp_calib_width,
lamda=args.lamda,
device=device,
comm=comm,
