Changeset 19dcb933 in sasmodels for README.rst

Timestamp:

Sep 3, 2014 3:16:10 AM (10 years ago)

Author:

Paul Kienzle <pkienzle@…>

Branches:

master, core_shell_microgels, costrafo411, magnetic_model, release_v0.94, release_v0.95, ticket-1257-vesicle-product, ticket_1156, ticket_1265_superball, ticket_822_more_unit_tests

Children:

1c7ffdc

Parents:

87985ca

Message:

build docs for models

File:

• README.rst (moved) (moved from sasmodels/models/README.rst) (3 diffs)

README.rst

@@ +1 @@
+Sasmodels
+=========
+
+Theory models for small angle scattering.
+
+The models provided are usable directly in the bumps fitting package and
+in the sasview analysis package.  If OpenCL is available, the models will
+run much faster.  If not, then precompiled versions will be included with
+the distributed package.  New models can be added if OpenCL or a C compiler
+is available.
+
 cylinder.c + cylinder.py is the cylinder model with renamed variables and
 sld scaled by 1e6 so the numbers are nicer.  The model name is "cylinder"
@@ -10 +21 @@
 
 Note: may want to rename form_volume to calc_volume and Iq/Iqxy to
-calc_Iq/calc_Iqxy. Similarly ER/VR go to calc_ER/calc_VR.
+calc_Iq/calc_Iqxy in model interface. Similarly ER/VR go to calc_ER/calc_VR.
 
 Note: It is possible to translate python code automatically to opencl, using
 something like numba, clyther, shedskin or pypy, so maybe the kernel functions
 could be implemented without any C syntax.
+
+Note: angular dispersion in theta is probably not calculated correctly, but
+is left this way for compatibility with sasview.
 
 Magnetism hasn't been implemented yet.  We may want a separate Imagnetic
@@ -23 +37 @@
 in the same data set.
 
-Need to write code to generate the polydispersity loops in python for
-kernels that are only implemented in python.  Also, need to implement
-an example kernel directly in python.
+Need to implement an example kernel directly in python.  Polydispersity
+loops should be generated automatically as they are for the OpenCL models.
+The kernels should be vectorized across Q.  These will need vectorized
+versions of numerical quadrature if we want to get reasonable performance.
+