v0.12.0

2023-05-14 (v 0.12.0) (MINOR release)

• Renamed NASimEnv.get_minimum_actions -> NASimEnv.get_minumum_hops to better reflect what it does (thanks @rzvnbr for the suggestion).

v0.11.0

2023-03-13 (v 0.11.0) (MINOR release)

• Migrated to gymnasium (formerly Open AI gym) fromOpen AI gym (thanks @rzvnbr for the suggestion).
• Fixed bug with action string representation (thanks @rzvnbr for the bug report)
• Added "sim to real considerations" explanation document to the docs (thanks @Tudyx for the suggestion)

v0.10.1

2023-02-27 (v 0.10.1) (MICRO release)

• Fixed bug for host based actions (thanks @nguyen-thanh20 for the bug report)

v0.10.0

2022-07-30 (v 0.10.0) (MINOR release)

• Fixed typos (thanks @francescoluciano)
• Updates to be compatible with latest version of OpenAI gym API (v0.25) (see Open AI gym API docs for details), notable changes include
  • Updated naming convention when initializing environments using the gym.make API (see gym load docs for details.)
  • Updated reset function to match new gym API (shouldn't break any implementations using old API)
  • Updated step function to match new gym API. It now returns two bools, the first specifies if terminal/goal state has been reached and the other specifies if the episode is terminated due to the scenario step limit (if any exists) has been reached. This change may break implementations and you may need to specify (or not) when initializing the gym environment using gym.make(env_id, new_step_api=True)

v0.9.1

2022-05-19 (v 0.9.1) (MICRO release)

• Fixed a few bugs and added some tests (thanks @simonsays1980 for the bug reports)

v0.9.0

2021-12-20 (v 0.9.0) (MINOR release)

• The value of a host is now observed when any level of access is gained on a host. This makes it so that agents can learn to decide whether to invest time in gaining root access on a host or not, depending on the host's value (thanks @jaromiru for the proposal).
• Initial observation of reachable hosts now contains the host's address (thanks @jaromiru).
• Added some support for custom address space bounds in when using scenario generator (thanks @jaromiru for the suggestion).

v0.8.0

2021-3-15 (v 0.8.0) (MINOR release)

• Added option of specifying a 'value' for each host when defining a custom network using the .YAML format. (Thanks to @Joe-zsc for the suggestion)
• Added the 'small-honeypot' scenario to included scenarios.

v0.7.5

2020-12-24 (v 0.7.5) (MICRO release)

• Added 'undefined error' to observation to fix issue with initial and later observations being indistinguishable.

v0.7.4

2020-12-17 (v 0.7.4) (MICRO release)

• Fixed issues with incorrect observation of host 'value' and 'discovery_value'. Now, when in partially observable mode, the agent will correctly only observe these values on the step that they are recieved.
• Some other minor code formatting fixes

v0.7.3

2020-09-23 (v 0.7.3) (MICRO release)

• Fixed issue with scenario YAML files not being included with PyPi package
• Added final policy visualisation option to DQN and Q-Learning agents