8. Wrappers

Sinergym has several wrappers in order to add some functionality in the environment that it doesn’t have by default. Currently, we have developed a normalization wrapper, multi-observation wrapper, multi-objective wrapper and Logger wrapper. The code can be found in sinergym/sinergym/utils/wrappers.py. You can implement your own wrappers inheriting from gym.Wrapper or some of its variants.

  • NormalizeObservation: It is used to transform observation received from simulator in values between 0 and 1.

  • LoggerWrapper: Wrapper for logging all interactions between agent and environment. Logger class can be selected in the constructor if other type of logging is required. For more information about Sinergym Logger visit Logger.

  • MultiObjectiveReward: Environment step will return a vector reward (selected elements in wrapper constructor, one for each objective) instead of a traditional scalar value. See #301.

  • MultiObsWrapper: Stack observation received in a history queue (size can be customized).

  • DatetimeWrapper: Wrapper to substitute day value by is_weekend flag, and hour and month by sin and cos values. Observation space is updated automatically.

  • PreviousObservationWrapper: Wrapper to add observation values from previous timestep to current environment observation. It is possible to select the variables you want to track its previous observation values.

  • DiscreteIncrementalWrapper: A wrapper for an incremental discrete setpoint action space environment. A discrete environment with only temperature setpoints control must be used with this wrapper.

Note

For examples about how to use these wrappers, visit Wrappers example.

Warning

The order of wrappers if you are going to use several at the same time is really important. The correct order is Normalization - Logger - Multi-Objectives - MultiObs and subsets (for example, Normalization - Multiobs is valid). It is recommended not to use several Wrappers whose action and/or observation space is modified at the same time.

Warning

If you add new observation variables to the environment than the default ones, you have to update the value range dictionary in sinergym/sinergym/utils/constants.py so that normalization can be applied correctly. Otherwise, you will encounter bug #249.