Contents of /release/5.2/doc/user/changes.tex

Revision 6694 - (show annotations)
Mon Jun 25 03:26:25 2018 UTC (7 months, 3 weeks ago) by jfenwick
File MIME type: application/x-tex
File size: 15752 byte(s)
Changelog


 1 2 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 3 % Copyright (c) 2003-2018 by The University of Queensland 4 5 % 6 % Primary Business: Queensland, Australia 7 % Licensed under the Apache License, version 2.0 8 9 % 10 % Development until 2012 by Earth Systems Science Computational Center (ESSCC) 11 % Development 2012-2013 by School of Earth Sciences 12 % Development from 2014 by Centre for Geoscience Computing (GeoComp) 13 % 14 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 15 16 \chapter{Changes from previous releases} 17 \label{app:changes} 18 19 \subsection*{5.2} 20 \begin{itemize} 21 \item Documentation improvements 22 \item Added makeTagMap function 23 \item Fixes to \texttt{tensor_transposed_mult} 24 \item Additional tests 25 \end{itemize} 26 27 \subsection*{5.1} 28 \begin{itemize} 29 \item Improvements to maths using complex values. 30 Note: Use the \texttt{promote()} method on Data objects (added in 5.0) to convert them to complex. 31 This does not change the value itself, (eg $7 \rightarrow 7+0j$) but matters for operations like \emph{sqrt} which behave differently with complex inputs. 32 \item Added \texttt{phase} operation to return the angular part/phase/arg of values. 33 \item Improved inversion. 34 \item Corrections to $\cdot\_transpose\_product$ operations. 35 \item Improved integration'' testing. 36 \item NetCDF support (if enabled), can now be built with support for netcdf-4 formatted files.\\ 37 See \texttt{scons/templates/README_FIRST} for the options. 38 \item Support for lazy expressions with complex values. 39 \item Solving PDEs with complex coefficients requires Trilinos support. 40 Since escript requires features which are not 41 compiled in by default, please consult the install guide if you wish to make use of 42 this feature. 43 \item Note that in the $4.2.0.1$ release, the license for the escript source code was changed to Apache2''. 44 This should be less restrictive than the previous license. 45 \end{itemize} 46 47 \subsection*{5.0} 48 \begin{itemize} 49 \item Minimum Python version is now 2.7. 50 \item Remaining code switched from C to C++ 51 \item Compilers are now required to support c++11 at minimum. 52 \item Some support for storing and manipulating complex values in Data objects. 53 \item Some solver configurations can solve PDEs with complex values. 54 \item Experimental support for Trilinos solvers and preconditioners. 55 \item \ripley by default no longer adjusts domain lengths when number of MPI ranks is unsuitable. A new function controls this behaviour. 56 \item Some C++ exceptions are now translated to equivalent Python exceptions (e.g. ValueError). 57 \item New \escript feature test function to interrogate build features at runtime. 58 \item Support for 'long' indices for very large local domains. 59 \item \escript is now more modular. New options can be used to select what to build. 60 \end{itemize} 61 62 \subsection*{4.0 to 4.1} 63 \begin{itemize} 64 \item Added multi-resolution \ripley domains 65 \item The gmshReader now supports reading with multiple processes 66 \item Using the help() function on some domains is now more informative 67 \item User guide updated with information on use of Dirac points 68 \item Minimizer misfit now available via the callback function 69 \item Specifying use of a direct solver without a direct solver being available now raises an exception rather than silently default to a non-direct solver 70 \item Synthetic seismic examples for various wave types now included in the examples distributed 71 \item Reading NetCDF files when the default value in the file is nan no longer causes issues 72 \item Various documentation updates and fixes in user guide, install guide, and code documentation 73 \item Various compatibility fixes with Python3 74 \item HTI and VTIWave models now support setting a q value for custom boundary conditions 75 \item Using a MPI-enabled gmsh with an MPI build of \escript should now reliably wait for gmsh to end 76 \item Added an example for voxet reading with \ripley 77 \end{itemize} 78 79 \subsection*{3.4.2 to 4.0} 80 \begin{itemize} 81 \item New spectral element domain, \speckley 82 \item CUDA-based solvers and block-diagonal system matrices in \ripley 83 \item DC Resistivity forward modeling 84 \item New build options: 85 \begin{itemize} 86 \item 'domains': allows only the specified domains to be built 87 \item 'prelaunch', 'launcher', 'postlaunch': allows build-time 88 customization of escript launcher 89 \item 'cuda', 'nvcc', 'nvccflags', 'thrust_prefix': path and flags for 90 CUDA compiler and thrust headers 91 \end{itemize} 92 \item Gmsh reader now \MPI parallelised 93 \item cycle and timestamp support for saveSilo and saveVTK 94 \item PDEs using \ripley now have per-PDE assemblers to allow domains to have 95 multiple PDEs of different types without conflict 96 \item General documentation updates, both guides and code commenting 97 \item Test framework compatible with older versions of Python 98 \item Removed BASHisms from \escript launcher 99 \item Data objects now have hasNaN() and replaceNaN() methods 100 \item readBinaryGrid() now supports scaling and 64-bit floats in non-native 101 byte order 102 \item Now compiles with Cray compiler 103 \item Wave models corrected 104 \item Equality checks on Data objects now throw exceptions 105 \item General changes to help compilers with \OPENMP parallelisation 106 \item Intel-based \OPENMP flavours now have proper thread binding 107 \item \ripley dirac points no longer get lost in domains with enormous 108 elements 109 \item lots of improvements in inversion toolbox \downunder and symbolic 110 toolbox 111 \end{itemize} 112 113 \subsection*{3.4.1 to 3.4.2} 114 \begin{itemize} 115 \item Changes to SolverOptions: Most of this is under the hood at this point 116 but any code which accesses parameters from a SolverOptions object like this: 117 \begin{python} 118 variable.getSolverOptions().DIRECT 119 \end{python} 120 will need to be modified to: 121 \begin{python} 122 SolverOptions.DIRECT 123 \end{python} 124 125 \item Data objects can be created using smoothed random data. 126 \item Support for OSX10.9 127 \item Support for additional problem types in downunder. 128 \item More efficient ripley domains for large compute clusters 129 \item Added symbolic toolbox and associated documentation 130 \item Ripley now supports Dirac delta functions. 131 \item Improved reading of netCDF files 132 \item Paso is now in its own namespace 133 \item Examples have been updated to use Python 3 behaviour (with future imports for Python 2). 134 \item Ripley supports reading from gzipped files. 135 \item Fixed fault in documentation generation 136 \item Fixes to keep clang++ happy 137 \end{itemize} 138 139 140 \subsection*{3.4 to 3.4.1} 141 \begin{itemize} 142 \item Renamed design.Design to design.AbstractDesign as a more explicit/descriptive name, {\bf this will break any existing custom implementation} until changed to match. 143 144 \item Efficiency improvements in esys.downunder inversions. 145 \item Implemented more CF conventions for the netCDF datasource and added GDAL support. 146 \item Strongly coupled joint inversion added. 147 \item Preliminary work on seismic modelling. 148 \item Added support for more input data types in ER Mapper files. 149 \item Improved stability for long MPI runs. 150 \item Added timeStepFormat option to \texttt{esys.escript.saveESD}. 151 \item Document generation process more stable. 152 \item Added support for smoothed random 2D data. 153 \item Documentation updated. 154 \item Improved read support for grid-reading. 155 \item Removed last usages of C compiler. 156 \item Efficiency improvements in \texttt{esys.downunder}. 157 \item Removed support for VSL random (\texttt{boost::random} is sufficient). 158 \item Fixes related to compatibility with Intel compiler. 159 \item Simplified module structure. 160 \item Various code cleanup and bug fixes. 161 \end{itemize} 162 163 \subsection*{3.3.1 to 3.4} 164 \begin{itemize} 165 \item This release does not use the support bundles from $3.0$ so building from source will be required in more cases. 166 \item In Debian and Ubuntu, there are now two packages. One containing escript itself and the other containing documentation. 167 This fits better with the way Debian does things. 168 \item esys.weipa can now write Voxet files for data on ripley. 169 \item All support for OpenDX has been dropped. 170 \item Improved documentation HTML API documentation. 171 \item Improvements to inversion module. 172 \end{itemize} 173 174 175 \subsection*{3.3 to 3.3.1} 176 \begin{itemize} 177 \item Source packages for Debian and Ubuntu can now be made directly by dpkg-source from the source tree\footnote{For this to work, you will need to place a file called \texttt{svn_version} containing the revision number in the root of the source.}. 178 \item non-uniform 1D interpolation (see Section~\ref{sec:interpolation}). 179 \item The minimum version of python required for escript is now $2.6$. 180 \item saveVTK and saveDX (previously deprecated) have been removed from the main escript module. 181 Please see the \module{weipa} documentation for export functionality. 182 \item The \module{downunder} inversion module is included. 183 \begin{itemize} 184 \item The documentation includes an inversion cookbook to get you started. 185 \item Most operations are possible with \escript's existing dependencies but some \emph{may} require \module{pyproj} or \module{gdal} 186 but \escript will let you know if that happens. 187 \end{itemize} 188 189 \end{itemize} 190 191 192 \subsection*{3.2.1 to 3.3} 193 \begin{itemize} 194 \item Experimental support for \pythonthree. 195 \item Parameter order for the table interpolation methods has changed to be consistent. 196 Please test your scripts if you use these functions. 197 \item setX on \finley and \dudley domains will now only accept coordinates from ContinuousFunction spaces. 198 This is to avoid some potentially nasty behaviour when using periodic boundary conditions. 199 You can still use setX, just make sure that you interpolate first. 200 \end{itemize} 201 202 203 \subsection*{3.1 to 3.2} 204 \begin{itemize} 205 \item The deprecated name for the launcher has been removed. To run scripts use 206 \emph{run-escript} not \emph{escript}. 207 \item \escript is no longer automatically imported by importing \finley. 208 You will need to import escript explicitly. (All of our example scripts do this anyway.) 209 \item An experimental version of the new Dudley domain is now available. 210 \item Various bug fixes and optimisations. 211 \item New algorithms for gmsh support. 212 \item Improvements to the AMG solver. AMG is the recommended solver for symmetric problems. 213 \item Fixed compilation issues using netcdf. 214 \item Redesigned configuration files to make it easier to compile from source 215 without finding the locations of all your libraries. 216 \item Faster rendering of documentation. 217 \item Documentation is now hyperlinked. 218 \item New data export module \weipa. The \texttt{saveVTK} functionality has 219 been moved into this module, and while calling \texttt{saveVTK} from the 220 \escript module still works it is discouraged and will be removed in a 221 future release. 222 \item New \texttt{esys.escript.DataManager} class for convenient checkpointing 223 and exporting of escript data. 224 \item \VisIt simulation interface for online data access and visualization. 225 \item Simpler interpolation and support for interpolation from 3D vectors. 226 \item HRZ lumping has been added and some clarification on how to use it. 227 \item Data objects populated with random'' values can be created. 228 \end{itemize} 229 230 \subsection*{3.0 to 3.1} 231 \begin{itemize} 232 \item The \emph{escript} launcher has been renamed to \emph{run-escript}. The 233 old name will still work in this version but will be removed in the future. 234 \item Lazy evaluation features have been improved and documented (see \Sec{sec:lazy}). 235 \item The \emph{escript} documentation now includes a new Cookbook which 236 demonstrates how to solve sample problems using escript. 237 \item Macro elements have been introduced. 238 \item The \var{saveDataCSV} method allows one or more \Data objects to be 239 exported in CSV format (see \Sec{sec:savedatacsv}). 240 \item \Data objects can be populated by interpolating from values in a table. 241 \item The new \var{getInfLocator} and \var{getSupLocator} functions in 242 \pdetools return Locators to a minimal/maximal point over the 243 data. 244 \item There is a new class to model fault systems (\var{esys.escript.faultsystems.FaultSystem}). 245 \item A beta version of an Algebraic Multigrid (AMG) solver is included. 246 \item Inverting square matrices larger than 3x3 is now permitted if escript is 247 compiled with Lapack support. 248 \item If escript is compiled with a modern compiler, then inf/sup/Lsup will now 249 report NaN, +/-inf as appropriate if those values appear in the data. 250 \item Data.setTags will take tag names as well as tag numbers. 251 \item The Scalar, Vector, Tensor, Tensor3, Tensor4 factory methods can now take 252 arrays/nested sequence like objects as their initial values. 253 \item escript.util.mkDir can now take a list of directories to create. 254 \item Behind the scenes, \PYTHON docstrings have been rewritten from epydoc to 255 restructured text. 256 \item Various other bug fixes and performance tweaks. 257 \end{itemize} 258 259 \subsection*{2.0 to 3.0} 260 \begin{itemize} 261 \item The major change here was replacing \module{numarray} with \numpy. 262 %% I've removed the link since it is no longer valid 263 % For general instructions on converting scripts to use numpy see\\ 264 265 The specific changes to \escript are: 266 \begin{itemize} 267 \item getValueOfDataPoint() which returned a \module{numarray}.array has been 268 replaced by getTupleForDataPoint() which returns a \PYTHON tuple 269 containing the components of the data point. In the case of matrices or 270 higher ranked data, the tuples will be nested. 271 Use \numpy.array(data.getTupleForDataPoint()) if a \numpyNDA object is 272 required. 273 \item getValueOfGlobalDataPoint() has similarly been replaced by 274 getTupleForGlobalDataPoint(). 275 \item \code{integrate(data)} now returns a \numpyNDA instead of a 276 \module{numarray}.array. 277 \end{itemize} 278 Any python methods which previously accepted \module{numarray} objects now 279 accept \numpy objects instead. 280 281 \item The way to define solver options for \LinearPDE objects has changed. 282 There is now a \SolverOptions object attached to the \LinearPDE 283 object which handles the options of solvers used to solve the PDE. 284 The following changes apply: 285 \begin{itemize} 286 \item The \method{setTolerance} and \method{setAbsoluteTolerance} methods 287 have been removed. Instead use \method{setTolerance} and 288 \method{setAbsoluteTolerance} on the \SolverOptions object. For example: 289 \method{getSolverOptions().setTolerance(\ldots)} 290 \item The \method{setSolverPackage} and \method{setSolverMethod} methods have 291 been removed. Instead use the methods \method{setPackage}, 292 \method{setSolverMethod} and \method{setPreconditioner}. For example: 293 getSolverOptions().setPackage(\ldots). 294 \item The static class variables defining packages, solvers and 295 preconditioners have been removed and are now accessed via the 296 corresponding static class variables in \SolverOptions. 297 For instance use \method{SolverOptions.PCG} instead of 298 \method{LinearPDE.PCG} to select the preconditioned conjugate gradient 299 method. 300 \item The \method{getSolution} now takes no argument. Use the corresponding 301 methods of the\\ \SolverOptions object returned by 302 \method{getSolverOptions()} to set values,\\ e.g. use 303 \method{getSolverOptions().setVerbosityOn()} instead of argument 304 \code{verbose=True} and \method{getSolverOptions().setIterMax(1000)} 305 instead of argument \code{iter_max=1000}. 306 \end{itemize} 307 \end{itemize} 308