You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Random comments on the poster (will add more to the issue as time goes by):
Mention how many different preprocessing passes there are
Maybe include two or so examples of preprocessing passes
API -> Context
Maybe have a list of classes that you implemented (PreprocessingPass, PreprocessingPassContext, PreprocessingPassRegistry) with a 1-2 sentence description of their purpose.
Drawing is missing PreprocessingPassRegistry
The text was updated successfully, but these errors were encountered:
"data structures like lists and bit vectors": replace "lists" with "arrays". Lists are not really natively supported
"An SMT, being a decision problem, can output two results": SMT what?
"exists a possible assignment" -> "exists an assignment"
"no combination of variable assignments" -> "no variable assignment"
"Prior to solving, the underbelly of CVC4 includes a preprocessing step is necessary to make equations readable to the solver or easier to solve": does not quite parse
"If-then-else clauses are unreadable" -> "If-then-else terms are not supported internally"
"before the DPLL solving" -> "before the DPLL(T) solving"
"Preprocessing Context – a class that receives variables from SmtEngine and passes them to the preprocessing passes": maybe something like: "Preprocessing Context – a restricted interface for preprocessing passes to interact with solver internals"
For the graphic: The SMTEngine owns both Context and Registry, no?
"The new implementation allow for more flexibility in certain test cases." -> "The new implementation allow for more flexibility."
"Ideally, the flexibility will allow for multiple ways of processing certain problems. In some circumstances, an ideal method of preprocessing will be found and can be used for similar SMT problems." -> "The flexibility can be used to automatically find better configurations by systematically trying different configurations on a fixed set of benchmarks."
Random comments on the poster (will add more to the issue as time goes by):
The text was updated successfully, but these errors were encountered: