Soil

Latest version: v0.20.8

Safety actively analyzes 681812 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 1 of 4

1.0.0

Version 1.0 will introduce multiple changes, especially on the `Simulation` class and anything related to how configuration is handled.
For an explanation of the general changes in version 1.0, please refer to the file `docs/notes_v1.0.rst`.

Added
* A modular set of classes for environments/models. Now the ability to configure the agents through an agent definition and a topology through a network configuration is split into two classes (`soil.agents.BaseEnvironment` for agents, `soil.agents.NetworkEnvironment` to add topology).
* Environments now have a class method to make them easier to use without a simulation`.run`. Notice that this is different from `run_model`, which is an instance method.
* Ability to run simulations using mesa models
* The `soil.exporters` module to export the results of datacollectors (`model.datacollector`) into files at the end of trials/simulations
* Agents can now have generators or async functions as their step or as states. They work similar to normal functions, with one caveat in the case of `FSM`: only time values (a float, int or None) can be awaited or yielded, not a state. This is because the state will not change, it will be resumed after the yield, at the appropriate time. To return to a different state, use the `delay` and `at` functions of the state.
* Simulations can now specify a `matrix` with possible values for every simulation parameter. The final parameters will be calculated based on the `parameters` used and a cartesian product (i.e., all possible combinations) of each parameter.
* Simple debugging capabilities in `soil.debugging`, with a custom `pdb.Debugger` subclass that exposes commands to list agents and their status and set breakpoints on states (for FSM agents). Try it with `soil --debug <simulation file>`
* The `agent.after` and `agent.at` methods, to avoid having to return a time manually.
Changed
* Configuration schema (`Simulation`) is very simplified. All simulations should be checked
* Model / environment variables are expected (but not enforced) to be a single value. This is done to more closely align with mesa
* `Exporter.iteration_end` now takes two parameters: `env` (same as before) and `params` (specific parameters for this environment). We considered including a `parameters` attribute in the environment, but this would not be compatible with mesa.
* `num_trials` renamed to `iterations`
* General renaming of `trial` to `iteration`, to work better with `mesa`
* `model_parameters` renamed to `parameters` in simulation
* Simulation results for every iteration of a simulation with the same name are stored in a single `sqlite` database

Removed
* The `time.When` and `time.Cond` classes are removed
* Any `tsih` and `History` integration in the main classes. To record the state of environments/agents, just use a datacollector. In some cases this may be slower or consume more memory than the previous system. However, few cases actually used the full potential of the history, and it came at the cost of unnecessary complexity and worse performance for the majority of cases.

0.20.8

Changed
* Tsih bumped to version 0.1.8
Fixed
* Mentions to `id` in docs. It should be `state_id` now.
* Fixed bug: environment agents were not being added to the simulation

0.20.7

Changed
* Creating a `time.When` from another `time.When` does not nest them anymore (it returns the argument)
Fixed
* Bug with time.NEVER/time.INFINITY

0.20.6

Fixed
* Agents now return `time.INFINITY` when dead, instead of 'inf'
* `soil.__init__` does not re-export built-in time (change in `soil.simulation`. It used to create subtle import conflicts when importing soil.time.
* Parallel simulations were broken because lambdas cannot be pickled properly, which is needed for multiprocessing.
Changed
* Some internal simulation methods do not accept `*args` anymore, to avoid ambiguity and bugs.

0.20.5

Changed
* Defaults are now set in the agent __init__, not in the environment. This decouples both classes a bit more, and it is more intuitive

0.20.4

Added
* Agents can now be given any kwargs, which will be used to set their state
* Environments have a default logger `self.logger` and a log method, just like agents

Page 1 of 4

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.