def create( spec, K=1, scheme = 'none', count = 20, *args ):
"""
@spec - Specification (size, endpoints, barriers); either exactly
specified in a file, or with numeric values in a list
@option_scheme - none|manual|optimal|small-world|random|ozgur's betweenness|ozgur's randomness|end
@n_actions - Number of steps that need to taken
comment : optimal(shortest path to destination)??|random|ozgur's betweenness|ozgur's randomness
"""
env = PlayRooms.create( spec, K )
# Percentage
if isinstance(count,str):
count = int(count[:-1])
count = count*env.S/100
# Add options for all the optimal states
O = []
if scheme == "none":
pass
elif scheme == "random-node":
O = OptionGenerator.optimal_options_from_random_nodes( env, count, *args )
elif scheme == "random-path":
O = OptionGenerator.optimal_options_from_random_paths( env, count, *args )
elif scheme == "betweenness":
O = OptionGenerator.optimal_options_from_betweenness( env, count, *args )
elif scheme == "small-world":
O = OptionGenerator.optimal_options_from_small_world( env, count, *args )
elif scheme == "betweenness+small-world":
O = OptionEnvironment.optimal_options_from_betweenness( env, count )
count_ = count - len( O )
O += OptionEnvironment.optimal_options_from_small_world( env, count_, *args )
elif scheme == "load":
O = OptionGenerator.options_from_file( count, *args )
else:
raise NotImplemented()
return OptionEnvironment( PlayRoomsOptions, env.S, env.A, env.P, env.R, env.R_bias, env.start_set, env.end_set, O )
def reset_rewards( env, spec, K=1, *args ):
O = env.O
env = PlayRooms.reset_rewards( env, spec, K )
return OptionEnvironment( PlayRoomsOptions, env.S, env.A, env.P, env.R, env.R_bias, env.start_set, env.end_set, O )
