Repository files navigation

AALpy is a light-weightautomata learning library written in Python.You can start learning models of black-box systems with a few lines of code.

Use the package managerpip to install the latest release of AALpy:

pip install aalpy

To install current version of the master branch (it might contain bugfixes and added functionalities between releases):

pip install https://github.com/DES-Lab/AALpy/archive/master.zip

The minimum required version of Python is 3.6.
Ensure that you haveGraphviz installed and added to your path if you want to visualize models.

For manual installation, clone the repo and installpydot (the only dependency).

High level introduction to automata learning can be found at the followingMedium post.If you are interested in automata learning or would like to understand the automata learning process in more detail,please check out ourWiki. On Wiki, you will find more detailed examples on how to use AALpy.

• https://github.com/DES-Lab/AALpy/wiki

Examples.py contains examples covering almost the whole of AALpy's functionality and its a great starting point.

All active automata learning procedures follow this high-level approach:

• Define the input alphabet and system under learning (SUL)
• Choose the equivalence oracle
• Run the learning algorithm

Passive learning algorithm simply require you to provide data in the appropriate format (check Wiki and Examples) and run the learning function.

fromaalpy.utilsimportload_automaton_from_file,generate_random_deterministic_automatafromaalpy.SULsimportAutomatonSULfromaalpy.oraclesimportRandomWalkEqOraclefromaalpy.learning_algsimportrun_Lstar,run_KV# load an automaton# automaton = load_automaton_from_file('path_to_the_file.dot', automaton_type='dfa')# or randomly generate onerandom_dfa=generate_random_deterministic_automata(automaton_type='dfa',num_states=8,input_alphabet_size=5,output_alphabet_size=2)# get input alphabet of the automatonalphabet=random_dfa.get_input_alphabet()# loaded or randomly generated automata are considered as BLACK-BOX that is queried# learning algorithm has no knowledge about its structure# create a SUL instance for the automaton/system under learningsul=AutomatonSUL(random_dfa)# define the equivalence oracleeq_oracle=RandomWalkEqOracle(alphabet,sul,num_steps=5000,reset_prob=0.09)# start learning# run_KV is for the most part reacquires much fewer interactions with the system under learninglearned_dfa=run_KV(alphabet,sul,eq_oracle,automaton_type='dfa')# or run L*# learned_dfa_lstar = run_Lstar(alphabet, sul, eq_oracle, automaton_type='dfa')# save automaton to file and visualize it# save_automaton_to_file(learned_dfa, path='Learned_Automaton', file_type='dot')# orlearned_dfa.save()# visualize automaton# visualize_automaton(learned_dfa)learned_dfa.visualize()# or just print its DOT representationprint(learned_dfa)

To make experiments reproducible, define a random seed at the beginning of your program.

fromrandomimportseedseed(2)# all experiments will be reproducible

AALpy has been used to:

• Learn Models of Bluetooth Low-Energy
• Find bugs in VIM text editor
• Learn Input-Output Behavior of RNNs
• Learn Models of GIT
• Solve RL Problems

If you use AALpy in your research, please cite us with of the following:

• Extended version (preferred)
• Tool paper

If you have research suggestions or you need specific help concerning your research, feel free to start adiscussion or contactedi.muskardin@silicon-austria.com.We are happy to help you and consult you in applying automata learning in various domains.

Pull requests are welcome. For significant changes, please open an issue first to discuss what you would like to change.In case of any questions or possible bugs, please open issues.