twodlearn.reinforce.systems module¶

class twodlearn.reinforce.systems.Acrobot(render_mode=None, output_dir=None)[source]¶

Bases: twodlearn.reinforce.systems.System

property initial_state[source]¶

class twodlearn.reinforce.systems.Cartpole(render_mode=None, output_dir=None)[source]¶

Bases: twodlearn.reinforce.systems.System

property initial_state[source]¶

class twodlearn.reinforce.systems.Cstr(dt=1.0, render_mode=None, output_dir=None)[source]¶

Bases: twodlearn.reinforce.systems.ModelicaSystem

property initial_state[source]¶

property n_actuators[source]¶

property n_sensors[source]¶

property n_states[source]¶

class twodlearn.reinforce.systems.ModelicaSystem(EnvClass, render_mode=None, output_dir=None)[source]¶

Bases: twodlearn.reinforce.systems.System

init_env(render_mode=None, output_dir=None)[source]¶

simulate(policy, steps, render=False)[source]¶

class twodlearn.reinforce.systems.System(EnvClass, render_mode=None, output_dir=None)[source]¶

Bases: object

property EnvClass[source]¶

do_render(env, mode='human')[source]¶

property dt[source]¶

init_env(render_mode=None, output_dir=None)[source]¶

property n_actuators[source]¶

property n_sensors[source]¶

property n_states[source]¶

sensor_model(x_k)[source]¶: calculates the sensor output from the system’s state

set_sensor_noise(std=0.0)[source]¶

simulate(policy, steps, render=False)[source]¶