Renamed WaveletRecon to L1WaveletRecon.

README.md

@@ -1,34 +1,34 @@
-``sigpy`` Overview
-================
+`Overview
+=========
 
 Introduction
 ------------
-
 ``sigpy`` is a Python package for signal reconstruction, with GPU support using ``cupy``.
 
 ``sigpy`` provides commonly used signal processing functions, including convolution, FFT, NUFFT, wavelet transform, and thresholding functions. All operations, except wavelet transform, can run on GPU. These operations are wrapped either in a linear operator class (``Linop``) or a proximal operator class (``Prox``) for easy usage in iterative algorithms. ``sigpy`` also implements popular iterative algorithms, such as conjugate gradient, (accelerated/proximal) gradient method, and primal dual hybrid gradient.
 
-``sigpy`` provides a submodule ``sigpy.mri`` that uses the core module to implement common MRI iterative reconstruction methods, including SENSE reconstruction, L1-wavelet reconstruction, and total-variation reconstruction. In addition, it provides convenient simulation and sampling functions, such as poisson-disk sampling function.
+``sigpy`` provides a submodule ``sigpy.mri`` that uses the core module to implement common MRI iterative reconstruction methods, including SENSE reconstruction, L1-wavelet reconstruction, total-variation reconstruction, and JSENSE reconstruction. In addition, it provides convenient simulation and sampling functions, such as poisson-disk sampling function.
 
 ``sigpy`` also provides a preliminary submodule ``sigpy.learn`` that implements convolutional sparse coding, and linear regression.
 
 Installation
 ------------
-The package is on PyPI, and can be installed via pip
+The package is on PyPI, and can be installed via pip:
 
 	pip install sigpy
-
-Alternatively, the package can also be installed with the following required packages.
 
-Requirements
-------------
-This package requires python3, numpy, scipy, pywavelets, and numba.
+For optional gpu support, the package requires ``cupy``.
 
-For optional gpu support, the package requires cupy.
+For optional distributed programming support, the package requires ``mpi4py``.
+
+Alternatively, the package can also be installed from source with the following requirements:
 
-For optional distributed programming support, the package requires mpi4py.
+- python3
+- numpy
+- scipy
+- pywavelets
+- numba
 
 Documentation
 -------------
-
 Our documentation is hosted on Read the Docs: https://sigpy.readthedocs.io

sigpy/mri/app.py

@@ -101,7 +101,7 @@ def __init__(self, ksp, mps, eps,
         super().__init__(A, ksp, self.img, proxg, eps, weights=weights, **kwargs)
 
 
-class WaveletRecon(sp.app.LinearLeastSquares):
+class L1WaveletRecon(sp.app.LinearLeastSquares):
     r"""L1 Wavelet regularized reconstruction.
 
     Considers the problem
@@ -145,7 +145,7 @@ def g(input):
         super().__init__(A, ksp, self.img, proxg=proxg, g=g, weights=weights, **kwargs)
 
 
-class WaveletConstrainedRecon(sp.app.L2ConstrainedMinimization):
+class L1WaveletConstrainedRecon(sp.app.L2ConstrainedMinimization):
     r"""L1 wavelet regularized constrained reconstruction.
 
     Considers the problem

sigpy/mri/app_test.py

@@ -47,23 +47,23 @@ def test_shepp_logan_SenseConstrainedRecon(self):
         img_rec = app.SenseConstrainedRecon(ksp, mps, std).run()
         npt.assert_allclose(img, img_rec, atol=1e-1, rtol=1e-1)
 
-    def test_shepp_logan_WaveletRecon(self):
+    def test_shepp_logan_L1WaveletRecon(self):
         img, mps, ksp = self.shepp_logan_setup()
         lamda = 0
 
-        img_rec = app.WaveletRecon(
+        img_rec = app.L1WaveletRecon(
             ksp, mps, lamda, alg_name='GradientMethod').run()
         npt.assert_allclose(img, img_rec, atol=1e-1, rtol=1e-1)
 
-        img_rec = app.WaveletRecon(ksp, mps, lamda, alg_name='PrimalDualHybridGradient',
-                                   max_iter=100).run()
+        img_rec = app.L1WaveletRecon(ksp, mps, lamda, alg_name='PrimalDualHybridGradient',
+                                     max_iter=100).run()
         npt.assert_allclose(img, img_rec, atol=1e-1, rtol=1e-1)
 
-    def test_shepp_logan_WaveletConstrainedRecon(self):
+    def test_shepp_logan_L1WaveletConstrainedRecon(self):
         img, mps, ksp = self.shepp_logan_setup()
         std = 0
 
-        img_rec = app.WaveletConstrainedRecon(ksp, mps, std).run()
+        img_rec = app.L1WaveletConstrainedRecon(ksp, mps, std).run()
         npt.assert_allclose(img, img_rec, atol=1e-1, rtol=1e-1)
 
     def test_shepp_logan_TotalVariationRecon(self):