kigen.transducer
Lossless machine learning: constructing a single symbol output transducer from input word, output symbol pairs by logic programming. In other words, constructing a Moore-machine. https://en.wikipedia.org/wiki/Moore_machine Internally both the states and the input symbols are represented as nonnegative integers, for the ease of handling by the logic engine through the finite domain package. State 0 is the initial state.
Lossless machine learning: constructing a single symbol output transducer from input word, output symbol pairs by logic programming. In other words, constructing a Moore-machine. https://en.wikipedia.org/wiki/Moore_machine Internally both the states and the input symbols are represented as nonnegative integers, for the ease of handling by the logic engine through the finite domain package. State 0 is the initial state.
checkclj
(check io-pairs {delta :delta omega :omega})Returns true if the given automaton (defined by solution, state transition function and output function) will indeed produce the output values given in the io-pairs. It uses format-flexible for processing the raw solution.
Returns true if the given automaton (defined by solution, state transition function and output function) will indeed produce the output values given in the io-pairs. It uses format-flexible for processing the raw solution.
check-fixedclj
(check-fixed io-pairs delta)Returns true if the given automaton (defined by delta, state transition function) will indeed produce the output values given in the io-pairs.
Returns true if the given automaton (defined by delta, state transition function) will indeed produce the output values given in the io-pairs.
fixed-output-transducerclj
(fixed-output-transducer io-pairs n)Given the the input-output pairs, and the number of states, this attempts to construct a suitable transducer. Both the inputs and the outputs are represented as nonnegative integers. The output is delta, the first is the transformation realized by input symbol 0, second for input symbol 1, and so on. It produces all solutions lazily, so if only a solution is needed, first can be used.
Given the the input-output pairs, and the number of states, this attempts to construct a suitable transducer. Both the inputs and the outputs are represented as nonnegative integers. The output is delta, the first is the transformation realized by input symbol 0, second for input symbol 1, and so on. It produces all solutions lazily, so if only a solution is needed, first can be used.
input-symbols-fnclj
(input-symbols-fn io-pairs)Returns all collected input symbols appearing in the input-output pairs without repetition. Returned as a vector, the indices can be used to refer to the symbols. The order of the symbols defined by the order of their appeareance in the io-pairs (through distinct).
Returns all collected input symbols appearing in the input-output pairs without repetition. Returned as a vector, the indices can be used to refer to the symbols. The order of the symbols defined by the order of their appeareance in the io-pairs (through distinct).
output-symbols-fnclj
(output-symbols-fn io-pairs)Returns all collected output symbols appearing in the input-output pairs without repetition. Returned as a vector, the indices can be used to refer to the symbols. The order of the symbols defined by the order of their appeareance in the io-pairs (through distinct).
Returns all collected output symbols appearing in the input-output pairs without repetition. Returned as a vector, the indices can be used to refer to the symbols. The order of the symbols defined by the order of their appeareance in the io-pairs (through distinct).
process-wordclj
(process-word delta initial-state input-word)Processes an input word (sequence of input symbols) by an automaton described by the delta state transition function (as vector of vectors) starting from the given initial state. It returns the resulting state.
Processes an input word (sequence of input symbols) by an automaton described by the delta state transition function (as vector of vectors) starting from the given initial state. It returns the resulting state.
process-wordoclj
(process-wordo delta initial-state input-word output)The relational version of process-word.
The relational version of process-word.
trajectoriesclj
(trajectories io-pairs {delta :delta omega :omega})Creates string representations of all trajectories by the io-pairs.
Creates string representations of all trajectories by the io-pairs.
trajectories-fixedclj
(trajectories-fixed io-pairs delta)Creates string representations of all trajectories by the io-pairs.
Creates string representations of all trajectories by the io-pairs.
trajectoryclj
(trajectory delta initial-state input-word)Processes an input word (sequence of input symbols) by an automaton described by the delta state transition function (as vector of vectors) starting from the given initial state. The whole trajectory (initital, all intermittent states and final state) is returned.
Processes an input word (sequence of input symbols) by an automaton described by the delta state transition function (as vector of vectors) starting from the given initial state. The whole trajectory (initital, all intermittent states and final state) is returned.
transducerclj
(transducer io-pairs n)Given the input-output pairs, and the number of states, this attempts to construct a suitable transducer. It produces all solutions lazily, so if only a solution is needed, first can be used.
Given the input-output pairs, and the number of states, this attempts to construct a suitable transducer. It produces all solutions lazily, so if only a solution is needed, first can be used.
cljdoc is a website building & hosting documentation for Clojure/Script libraries
× close