Conversation
…to avoid error messages after calling them
…d to ensure the set item is defined as tuple
… of the selection. The topology would be empty if it is not defined.
…rdf is running without error messages
Feature: length __len__ in AtomicSystem
|
I really like the idea of introducing the "None" if I get it right. Could you provide a sample inputfile? And you execute it then just with |
Yes I just use The input file looks like: |
| self.logger.error( | ||
| ( | ||
| "The restart_file key or the moldescriptor_file key with restart_file key " | ||
| "have to be set in order to use the RDF analysis." |
There was a problem hiding this comment.
If I recall it correctly, this does not have to be the case. For example if you have a plain xyz trajectory and want to give just the plain indices than this should be fine without any restart file or am I wrong?
The restart file should only be necessary if the user want to have a special selection based on atomtypes, molecules, ...
maybe this does not work as expected atm, but then we should tackle the problem there. So that if only an index selection is given the RDF calculation works without restart file.
There was a problem hiding this comment.
Okay so i went through your description once again and I see you already mention this issue. Before merging this we should find a solution on how to calculate RDF's without a rst file.
|
The tests are blocked most likely due to a numpy function |
…d to ensure the set item is defined as tuple
… of the selection. The topology would be empty if it is not defined.
…rdf is running without error messages
…into bugfix/rdf_bugfix
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## dev #114 +/- ##
==========================================
+ Coverage 85.43% 86.84% +1.40%
==========================================
Files 125 125
Lines 5027 5047 +20
==========================================
+ Hits 4295 4383 +88
+ Misses 732 664 -68
|
Co-authored-by: Jakob Gamper <97gamjak@gmail.com>
…into bugfix/rdf_bugfix
tests/io/inputFileReader/PQAnalysis/test_pqanalyisis_input_file_reader.py
Outdated
Show resolved
Hide resolved
Tests for TestRDFDataWriter and TestRDFLogWriter.
Error in test_write_before_run:
"Reference selection: {rdf.reference_selection}"
has in the output additional curly brackets:
"Reference selection: {{rdf.reference_selection}}"
Suggestion to remove the additional brackets in rdfOutputFileReader
if these brackets are not indendet.
Further change:
additional adding to .gitignore for the .coverage.* on Mac
- Initialization is checked by coveraging also the if conditions - average_volume() function is checked
|
Does it now work also for plain |
|
``> Does it now work also for plain I changed the topology definition in the
Now it is working without throwing an error and the comparison with restart file topology gives the same rdf results. |
|
PYLINT REPORT Your code has been rated at 9.79/10 (previous run: 9.79/10, -0.00) Full reportRaw metrics
Duplication
Messages by category
% errors / warnings by module
Messages
|
3 participants
Bug in
rdfIf not all optional keys are defined in the input file, a
PQKeyErroroccurs. But not all optional keys can be set simultaneously.Suggestion to solve it:
This was solved by include all not defined optional keys in the dictionary and set them at
NoneinRDFInputFileReaderbefore the keys are actually used.Further, for the parsing also
NoneTypein_parse.pythe conditions for theInputFileErrormessages are changed so thatNoneis not throwing an error anymore.The
rdfprogram works only if a topology is given either withrestart_fileor (moldescriptor_fileandrestart_file).If it is not given, an error occurs during the selection part of the target and reference molecules
Therefore, in
RDFInputFileReaderan error message was included.The
target_densityinRDFwas calculated:but this is only defined if
no_intra_molecular=True. So this was changed to:At the moment, no unexpected errors occur.