1 This version of PsN is tested with perl-5.8.4
2 Depending on your distribution you may have to install the following
8 All available from CPAN ( www.cpan.org )
11 ----------------------------------------------
13 Changes from 2.2.0 to 2.2.1 ( update for PsN course )
21 - Removed unecessary regular expression in output file parser.
23 - Not really a bug. But i tried to add "FIX" as a record option to a theta
24 records. Then when "store_init" was called on the "FIX" option PsN crashed
25 because the option was not a "theta_option". So i added a check for
26 "store_init" abilties... if FIX would become a record option in the future.
28 - Member accessor "datafile" replaced with "datafiles" in model. Affected
29 only the bootstrap which have been updated.
31 - An assumptions about outputfiles crashed sumo occationally. Some more
32 rigorous contentchecking has been added and fixed the problem.
34 Changes from 2.1.10 to 2.2.0 ( released for PsN course )
38 - Automatic incrementation of directory names are a bit smarter (no
39 more "directory modelfit_dirX is in the way" messages).
41 - Clean is much more stringent and intelligent and now manages to
42 remove entire tool directories if desired.
44 - option list printed from "-h|-?" only shows utility specific options.
46 - Default values configurable on a per utility basis.
48 - Online documentation improved
50 - Parallel run of NONMEM is not aborted as before if a outputfile is
51 not correctly parsed. It is just ignored, as if it had not
52 minimized successfully. (actually it is the output parser that
53 doesn't die, but the effect is the same)
55 - Parallel run is a bit nicer to the system when compiling many
56 NONMEMS. By adding a configurable delay between starts.
58 - Matlab and R scripts for various plots from cdd and bootstrap
59 results are distributed with PsN.
61 - Vast improvments in results presentations from all tools.
63 - Improvements in rerun algoritm with new "min_retries" option.
65 - Output from running utilties are much nicer.
67 - new "-quick_summary" prints minimization message and ofv value for
68 each finished run. As does "-verbose" option which also prints
69 "R:X" for each retrie (X is the retrie number).
71 - new "-summary" options prints diagnostics about the tool run when
72 it is finished. For "execute" the "-summary" option will have the
73 same function as the "sumo" utiltiy as well as some diagnostics.
75 - PsN now recognizes that control files may require fortran files and
76 copies them to the run directory. No need to add fotran files to
77 the --extra_files option any more.
79 - Added --extra_output which allows you to copy more than the output
80 file from the run directory.
83 - Using low_cutoff and hi_cutoff in data::sd resulted in PsN trying to use
84 the new compact data strings as the old arrays. fixed.
88 - Cleanup of dubious utility options.
89 - Option help sorted alphabetically.
90 - "flush", "target" removed from model.
91 - "remove_temp_files" removed from modelfit.
92 - Changed the Makefile considerably, much simpler now.
93 - added "column_to_array" function in data module.
95 Changes from 2.1.9 to 2.1.10
98 - The 'picky' option of all utilities now triggers on
99 COVARIANCE STEP ABORTED and
100 PROGRAM TERMINATED BY OBJ
102 Changes from 2.1.8 to 2.1.9
105 - Nordugrid support reintroduced (but will probably not be developed further)
106 - Sumo now supports multiple problems and subproblems.
107 - Sumo has new option "precision" which control the number of digits
108 to present estimates with.
109 - An extra 'short' logfile is output from the scm with the most important
113 - Bootstrap option "stratify_on" doesn't work with columns defined in
114 extra data files given with "extra_data" option. Not Fixed.
115 - Resumes of scm's with extra_data_files is now possible. Still some
116 problems with scm's where search_direction=both.
117 - LSF options was not communicated properly between tools and subtools
118 so submission of bootstraps, llp, cdd and scm did not work. Fixed.
119 - Default values for LSF can now be configured in psn.conf.
120 - Print order of modelfiles places $MIX before $PRED. Fixed.
121 - extra_files option was broken. Fixed.
124 - Tool will always fork subtools, this will save memory in the scm (and
125 other recursive tools).
126 - Common options need only be specified once. In one of three categories:
127 model_options, tool_options, script_options.
128 - models can now be created with "reference" objects. This solved the
130 - Added function to model that cleans extra data code.
132 Changes from 2.1.7 to 2.1.8
135 - Rudimentary support for execution on Platform LSF grid systems is now
136 included. The following extra attributes are available to all tools
137 for controlling the execution:
140 lsf_queue - Use this queue on the grid
141 lsf_options - General options for the grid queuing system.
142 lsf_project_name - Optional. May be used by the grid system
143 administrator to register control project
145 lsf_job_name - Optional. May be used by a grid system
146 administrator to control multiple runs on a
149 - All tools now have a raw_results file holding the diagnotics and
150 parameter estimates for each NONMEM run.
151 - All tools now handles model files containing one $PROBLEM and no
153 - The Bootstrap and CDD routines now create separate files holding
154 information about which individuals that were included/excluded
155 in/from each created data set.
156 - The Bootstrap results file now has warnings section to better
157 display which runs that were used for result calculations and which
159 - PsN now tries to read psn.conf from the users home directory. (Will
160 require some extra documentation).
161 - Optimized modelfit thread operation by removing Storable::Store.
162 - The clean options has four levels of operation.
164 0 - No cleaning at all.
165 1 - Clean more, but make sure resumes work (this is default).
166 2 - Clean more, but make sure summarize work.
167 3 - Clean everything, keeping only results and logs.
169 - Some attributes and accessors of the PsN-Core class 'output' have
170 had their names changed to make it more clear what their purposes
172 termination is now called minimization_successful
173 termination_message is now called minimization_message
174 covstep_termination is now called covariance_step_successful
176 these attributes are new:
178 covariance_step_warnings (boolean; 0 or 1)
179 estimate_near_boundary (boolean; 0 or 1)
181 The attribute which was previously called termination could have the
182 values 0, 0.5 and 1 for 'minimization terminates', 'minimization
183 successful but warnings or errors were printed from the covariance
184 step' and finally 'minimization successful'. The new attribute
185 'minimization_successful' can only take the values 0 and 1 and the
186 new attribute covariance_step_warnings now takes the value 1 when
187 covariance step warnings were found. Checks are also perfomed to
188 test whether the estimates are close to any boundary.
189 - Html documents is now generated from command line help text.
190 which ensures that they are consistent and up to date.
191 - Personal configuration file now available under Unix like OS:s
194 - Options specific to modelfiles were ignored in some Utilties.
195 - Gradients now output in windows. Just like with nmfe.
196 - Bootstrap now consumes considerably less memory.
197 - Removed unused/obsolete "resume" function from modelfit.
198 - Sorted method entries in modelfit diagram.
199 - Fixed bugs in parsing of OFV and eigenvalues.
201 Changes from 2.1.6 to 2.1.7
204 - "clean" and "compress" options now work under Windows.
205 - SCM now handles crashed runs appropriately.
206 - The ofv_forward, ofv_backward, p_forward and p_backward options of
207 scm now works as they should.
210 - SCM improvements. The final model is now available in the
211 first-level scm directory of each step.
212 - The Utilties -help option is improved. You can now give a list of
213 options to -help for specific documentation of those options, like
216 $ scm -help do_not_drop
218 - Minor changes in how the [nm_version] section in psn.conf i
219 handled. Look in the psn.conf file that is distributed with PsN for
221 - SDE modelfiles are now supported. (see the -sde_records option)
222 - The update_inits method of the model class now accepts an output
224 - Some attributes and accessors of the PsN-Core class 'output' have
225 had their names changed to make it more clear what their purposes
227 finalgrad is now called final_gradient
228 parampath is now called parameter_path
229 gradientpath is now called gradient_path
230 covterm is now called covstep_termination
231 - The Utilties now saves the command line that started it in a file
232 "command.txt" under the corresponding directory.
233 - Added a "silent" option to the Utilties that quenches all output.
234 - "nice" option added to utilties that allows you to set priority in
236 - Rudimentary support for NorduGrid reinstated.
237 - Installer script rewritten. It now supports Perl installations in
238 nonstandard directories. Discrepancies between *nix and MSWin32 is
242 - The code that is autogenerated with "dia2code" is optimized a
243 bit and saves a hash "__valid_parameters" in each object, which
244 allows the constructors to be used as a generic (shallow) copy.
247 - continuous spelled correctly throughout SCM.
250 Changes from 2.1.5 to 2.1.6
253 - Stratify_on bug in bootstrap method fixed.
254 - Data files in directories different from model file are now
256 - Path name handling reimplemented. Relative paths are now handled
258 - Models that generate table files would make PsN complain about
260 - Its now possible to specify multiple output names as a
261 comma-separated list to execute.
262 - The scm can now go backwards.
263 - Handling of table files now responds to the
264 "ignore_missing_output_files" flag.
265 - nmtran error messages was lost in Windows 2K. Execution of nmtran
266 is more plattform independent now. It uses a temporary file, but
267 it is never bigger than the nmtran messages.
268 - Fixed a bug that limited choice of compiler under unix.
269 - Typo in scm utility fixed. (included_relations in backwards step).
272 - Added support for "$NONPARAMETRIC" record in model files.
273 - A few improved error messages related to parsing of model files.
274 - The SCM config file now features sections for parameter bounds and
278 - psn.conf settings are now available as global variables. psn.conf