chemistrylab.benches

Characterization Bench

class chemistrylab.benches.characterization_bench.CharacterizationBench(observation_list, targets, n_vessels)

Bases: object

A set of methods made available to inspect an inputted vessel.

Parameters:

• observation_list (Tuple[str]) – Ordered list of observations to make (see Method Map)

• targets (Tuple[str]) – A list of target materials

• n_vessels (int) – The (maximum) number of vessels included in an observation

Method Map:

Key	Method
‘spectra’	get_spectra
‘layers’	get_layers
‘targets’	encode_target
‘PVT’	encode_PVT

encode_PVT(vessel: Vessel)

Returns:

a size 3 array containing [temperature,volume,pressure]

Return type:

np.array

encode_target(vessel: Vessel)

Returns:

a 1D one-hot encoding of the target material (note self.target must be set before doing this)

Return type:

np.array

get_layers(vessel: Vessel)

Returns:

a 1D array of vessel layer information

Return type:

np.array

get_observation(vessels: Vessel, target: str)

Returns a concatenation of observations of the vessels provided, using the list of observations provided in __init__

Parameters:

• vessels (Tuple[Vessel]) – A list of vessels you want an observtation of.

• target (str) – The current target material

get_observation_subset(vessels, subset: set, target: str = '')

get_spectra(vessel: Vessel, materials: Tuple[Material] | None = None, overlap: bool = True)

Class method to generate total spectral data using a gaussian decay.

Parameters:

• vessel (Vessel) – The vessel inputted for spectroscopic analysis.

• materials (Optional[Tuple[material.Material]]) – List of materials to get the spectra from

• overlap (bool) – Indicates if the spectral plots show overlapping signatures. Defaults to False.

Returns:

A 1D array containing the absorption data of the present materials.

Return type:

np.array

chemistrylab.benches.characterization_bench.calc_absorb3(item, C, x, w_min, w_max, absorb)

Distillation Bench

class chemistrylab.benches.distillation_bench.GeneralWurtzDistill_v2

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action	Vessel	Event Performed	Event Parameters	Other Vessel
0	boil_vessel	heat contact	Tf: 270.0, ht: 200.0	Beaker 1
1	boil_vessel	heat contact	Tf: 277.5, ht: 200.0	Beaker 1
2	boil_vessel	heat contact	Tf: 285.0, ht: 200.0	Beaker 1
3	boil_vessel	heat contact	Tf: 292.5, ht: 200.0	Beaker 1
4	boil_vessel	heat contact	Tf: 300.0, ht: 200.0	Beaker 1
5	boil_vessel	heat contact	Tf: 370.0, ht: 30.0	Beaker 1
6	boil_vessel	heat contact	Tf: 527.5, ht: 30.0	Beaker 1
7	boil_vessel	heat contact	Tf: 685.0, ht: 30.0	Beaker 1
8	boil_vessel	heat contact	Tf: 842.5, ht: 30.0	Beaker 1
9	boil_vessel	heat contact	Tf: 1000.0, ht: 30.0	Beaker 1
10	boil_vessel	pour by volume	volume: 0.0	Beaker 1
11	boil_vessel	pour by volume	volume: 0.111	Beaker 1
12	boil_vessel	pour by volume	volume: 0.222	Beaker 1
13	boil_vessel	pour by volume	volume: 0.333	Beaker 1
14	boil_vessel	pour by volume	volume: 0.444	Beaker 1
15	boil_vessel	pour by volume	volume: 0.556	Beaker 1
16	boil_vessel	pour by volume	volume: 0.667	Beaker 1
17	boil_vessel	pour by volume	volume: 0.778	Beaker 1
18	boil_vessel	pour by volume	volume: 0.889	Beaker 1
19	boil_vessel	pour by volume	volume: 1.0	Beaker 1
20	Beaker 1	pour by volume	volume: 0.0	Beaker 2
21	Beaker 1	pour by volume	volume: 0.111	Beaker 2
22	Beaker 1	pour by volume	volume: 0.222	Beaker 2
23	Beaker 1	pour by volume	volume: 0.333	Beaker 2
24	Beaker 1	pour by volume	volume: 0.444	Beaker 2
25	Beaker 1	pour by volume	volume: 0.556	Beaker 2
26	Beaker 1	pour by volume	volume: 0.667	Beaker 2
27	Beaker 1	pour by volume	volume: 0.778	Beaker 2
28	Beaker 1	pour by volume	volume: 0.889	Beaker 2
29	Beaker 1	pour by volume	volume: 1.0	Beaker 2
30	N/A	End Experiment	N/A	N/A

metadata: dict[str, Any] = {'render_fps': 10, 'render_modes': ['rgb_array']}

class chemistrylab.benches.distillation_bench.WurtzDistillDemo_v0

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action	Vessel	Event Performed	Event Parameters	Other Vessel
0	boil_vessel	heat contact	Tf: 300, ht: 20	Beaker 1
1	boil_vessel	heat contact	Tf: 270, ht: 20	Beaker 1
2	boil_vessel	heat contact	Tf: 1000, ht: 3	Beaker 1
3	boil_vessel	pour by volume	volume: 0.02	Beaker 1
4	Beaker 1	pour by volume	volume: 0.02	Beaker 2
5	N/A	End Experiment	N/A	N/A

get_keys_to_action()

metadata: dict[str, Any] = {'render_fps': 60, 'render_modes': ['rgb_array']}

chemistrylab.benches.distillation_bench.wurtz_vessel(add_mat)

Function to generate an input vessel for the wurtz distillation experiment.

Parameters:

add_mat (str) – The target material to include in the vessel

Returns:

A vessel containing add_mat and some undesired materials

Return type:

Vessel

Extraction Bench

class chemistrylab.benches.extract_bench.GeneralWurtzExtract_v2

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action	Vessel	Event Performed	Event Parameters	Other Vessel
0	extraction_vessel	drain by pixel	n_pixel: 2	Beaker 1
1	extraction_vessel	drain by pixel	n_pixel: 4	Beaker 1
2	extraction_vessel	drain by pixel	n_pixel: 6	Beaker 1
3	extraction_vessel	drain by pixel	n_pixel: 8	Beaker 1
4	extraction_vessel	drain by pixel	n_pixel: 10	Beaker 1
5	extraction_vessel	mix	t: -0.2	N/A
6	extraction_vessel	mix	t: -0.4	N/A
7	extraction_vessel	mix	t: -0.6	N/A
8	extraction_vessel	mix	t: -0.8	N/A
9	extraction_vessel	mix	t: -1.0	N/A
10	Beaker 1	pour by volume	volume: 0.2	extraction_vessel
11	Beaker 1	pour by volume	volume: 0.4	extraction_vessel
12	Beaker 1	pour by volume	volume: 0.6	extraction_vessel
13	Beaker 1	pour by volume	volume: 0.8	extraction_vessel
14	Beaker 1	pour by volume	volume: 1.0	extraction_vessel
15	Beaker 2	pour by volume	volume: 0.2	extraction_vessel
16	Beaker 2	pour by volume	volume: 0.4	extraction_vessel
17	Beaker 2	pour by volume	volume: 0.6	extraction_vessel
18	Beaker 2	pour by volume	volume: 0.8	extraction_vessel
19	Beaker 2	pour by volume	volume: 1.0	extraction_vessel
20	extraction_vessel	pour by volume	volume: 0.2	Beaker 2
21	extraction_vessel	pour by volume	volume: 0.4	Beaker 2
22	extraction_vessel	pour by volume	volume: 0.6	Beaker 2
23	extraction_vessel	pour by volume	volume: 0.8	Beaker 2
24	extraction_vessel	pour by volume	volume: 1.0	Beaker 2
25	C6H14 Vessel	pour by volume	volume: 0.1	extraction_vessel
26	C6H14 Vessel	pour by volume	volume: 0.2	extraction_vessel
27	C6H14 Vessel	pour by volume	volume: 0.3	extraction_vessel
28	C6H14 Vessel	pour by volume	volume: 0.4	extraction_vessel
29	C6H14 Vessel	pour by volume	volume: 0.5	extraction_vessel
30	diethyl ether Vessel	pour by volume	volume: 0.1	extraction_vessel
31	diethyl ether Vessel	pour by volume	volume: 0.2	extraction_vessel
32	diethyl ether Vessel	pour by volume	volume: 0.3	extraction_vessel
33	diethyl ether Vessel	pour by volume	volume: 0.4	extraction_vessel
34	diethyl ether Vessel	pour by volume	volume: 0.5	extraction_vessel
35	extraction_vessel, Beaker 1, Beaker 2	mix	t: 0.01	N/A
36	extraction_vessel, Beaker 1, Beaker 2	mix	t: 0.02	N/A
37	extraction_vessel, Beaker 1, Beaker 2	mix	t: 0.04	N/A
38	extraction_vessel, Beaker 1, Beaker 2	mix	t: 0.08	N/A
39	extraction_vessel, Beaker 1, Beaker 2	mix	t: 0.16	N/A
40	N/A	End Experiment	N/A	N/A

metadata: dict[str, Any] = {'render_fps': 10, 'render_modes': ['rgb_array']}

class chemistrylab.benches.extract_bench.SeparateTest_v0

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action	Vessel	Event Performed	Event Parameters	Other Vessel
0	extraction_vessel	drain by pixel	n_pixel: 1	Beaker 1
1	extraction_vessel	mix	t: -0.02	N/A
2	Beaker 1	pour by volume	volume: 0.02	extraction_vessel
3	Beaker 2	pour by volume	volume: 0.02	extraction_vessel
4	extraction_vessel	pour by volume	volume: 0.02	Beaker 2
5	C6H14 Vessel	pour by volume	volume: 0.01	extraction_vessel
6	diethyl ether Vessel	pour by volume	volume: 0.01	extraction_vessel
7	H2O Vessel	pour by volume	volume: 0.01	extraction_vessel
8	H2O Vessel	pour by volume	volume: 0.01	extraction_vessel
9	ethyl acetate Vessel	pour by volume	volume: 0.01	extraction_vessel
10	extraction_vessel, Beaker 1, Beaker 2	mix	t: 0.001	N/A
11	extraction_vessel, Beaker 1, Beaker 2	mix	t: 0.016	N/A
12	N/A	End Experiment	N/A	N/A

metadata: dict[str, Any] = {'render_fps': 60, 'render_modes': ['rgb_array']}

step(action)

Here, actions are performed by ‘_perform_discrete_action’ or ‘_perform_continuous_action’. Afterwards, all vessels in react_list have their concentrations updated by a reaction (specified in __init__). Finally, a CharacterizationBench is used to generate an observation, and a reward function provides a reward (if a terminal state was reached).

Parameters:

action (int or 1D array) – The action to be performed

class chemistrylab.benches.extract_bench.WaterOilExtract_v0

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action	Vessel	Event Performed	Event Parameters	Other Vessel
0	extraction_vessel	drain by pixel	n_pixel: 2	Beaker 1
1	extraction_vessel	drain by pixel	n_pixel: 4	Beaker 1
2	extraction_vessel	drain by pixel	n_pixel: 6	Beaker 1
3	extraction_vessel	drain by pixel	n_pixel: 8	Beaker 1
4	extraction_vessel	drain by pixel	n_pixel: 10	Beaker 1
5	extraction_vessel	mix	t: -0.2	N/A
6	extraction_vessel	mix	t: -0.4	N/A
7	extraction_vessel	mix	t: -0.6	N/A
8	extraction_vessel	mix	t: -0.8	N/A
9	extraction_vessel	mix	t: -1.0	N/A
10	Beaker 1	pour by volume	volume: 0.2	extraction_vessel
11	Beaker 1	pour by volume	volume: 0.4	extraction_vessel
12	Beaker 1	pour by volume	volume: 0.6	extraction_vessel
13	Beaker 1	pour by volume	volume: 0.8	extraction_vessel
14	Beaker 1	pour by volume	volume: 1.0	extraction_vessel
15	Waste Vessel	pour by volume	volume: 0.2	extraction_vessel
16	Waste Vessel	pour by volume	volume: 0.4	extraction_vessel
17	Waste Vessel	pour by volume	volume: 0.6	extraction_vessel
18	Waste Vessel	pour by volume	volume: 0.8	extraction_vessel
19	Waste Vessel	pour by volume	volume: 1.0	extraction_vessel
20	extraction_vessel	pour by volume	volume: 0.2	Waste Vessel
21	extraction_vessel	pour by volume	volume: 0.4	Waste Vessel
22	extraction_vessel	pour by volume	volume: 0.6	Waste Vessel
23	extraction_vessel	pour by volume	volume: 0.8	Waste Vessel
24	extraction_vessel	pour by volume	volume: 1.0	Waste Vessel
25	C6H14 Vessel	pour by volume	volume: 0.1	extraction_vessel
26	C6H14 Vessel	pour by volume	volume: 0.2	extraction_vessel
27	C6H14 Vessel	pour by volume	volume: 0.3	extraction_vessel
28	C6H14 Vessel	pour by volume	volume: 0.4	extraction_vessel
29	C6H14 Vessel	pour by volume	volume: 0.5	extraction_vessel
30	H2O Vessel	pour by volume	volume: 0.1	extraction_vessel
31	H2O Vessel	pour by volume	volume: 0.2	extraction_vessel
32	H2O Vessel	pour by volume	volume: 0.3	extraction_vessel
33	H2O Vessel	pour by volume	volume: 0.4	extraction_vessel
34	H2O Vessel	pour by volume	volume: 0.5	extraction_vessel
35	extraction_vessel, Beaker 1, Waste Vessel	mix	t: 0.01	N/A
36	extraction_vessel, Beaker 1, Waste Vessel	mix	t: 0.02	N/A
37	extraction_vessel, Beaker 1, Waste Vessel	mix	t: 0.04	N/A
38	extraction_vessel, Beaker 1, Waste Vessel	mix	t: 0.08	N/A
39	extraction_vessel, Beaker 1, Waste Vessel	mix	t: 0.16	N/A
40	N/A	End Experiment	N/A	N/A

metadata: dict[str, Any] = {'render_fps': 10, 'render_modes': ['rgb_array']}

class chemistrylab.benches.extract_bench.WurtzExtractDemo_v0

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action	Vessel	Event Performed	Event Parameters	Other Vessel
0	extraction_vessel	drain by pixel	n_pixel: 1	Beaker 1
1	extraction_vessel	mix	t: -0.02	N/A
2	Beaker 1	pour by volume	volume: 0.02	extraction_vessel
3	Beaker 2	pour by volume	volume: 0.02	extraction_vessel
4	extraction_vessel	pour by volume	volume: 0.02	Beaker 2
5	C6H14 Vessel	pour by volume	volume: 0.01	extraction_vessel
6	diethyl ether Vessel	pour by volume	volume: 0.01	extraction_vessel
7	extraction_vessel, Beaker 1, Beaker 2	mix	t: 0.001	N/A
8	extraction_vessel, Beaker 1, Beaker 2	mix	t: 0.016	N/A
9	N/A	End Experiment	N/A	N/A

get_keys_to_action()

metadata: dict[str, Any] = {'render_fps': 60, 'render_modes': ['rgb_array']}

chemistrylab.benches.extract_bench.make_solvent(mat)

Makes a Vessel with a single material

chemistrylab.benches.extract_bench.oil_vessel()

Function to generate an input vessel for the oil extraction bench.

Returns: - extract_vessel (Vessel): A vessel containing oil and NaCl

chemistrylab.benches.extract_bench.test_vessel(x)

chemistrylab.benches.extract_bench.wurtz_vessel(add_mat='')

Function to generate an input vessel for the wurtz extraction bench.

Args: - add_mat (str): The target material to include in the vessel

Returns: - extract_vessel (Vessel): A vessel containing add_mat and some undesired materials

Reaction Bench

class chemistrylab.benches.reaction_bench.FictReactBandit_v0(targets=None)

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action Index	Vessel	Event Performed	Event Parameters	Other Vessel
0	Reaction Vessel	heat contact	Tf: [273,373], ht: 300	Reaction Vessel
1	fict_A Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
2	fict_B Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
3	fict_C Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
4	fict_D Vessel	pour by percent	fraction: [0,1]	Reaction Vessel

step(action)

Here, actions are performed by ‘_perform_discrete_action’ or ‘_perform_continuous_action’. Afterwards, all vessels in react_list have their concentrations updated by a reaction (specified in __init__). Finally, a CharacterizationBench is used to generate an observation, and a reward function provides a reward (if a terminal state was reached).

Parameters:

action (int or 1D array) – The action to be performed

class chemistrylab.benches.reaction_bench.FictReactDemo_v0

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action Index	Vessel	Event Performed	Event Parameters	Other Vessel
0	Reaction Vessel	heat contact	Tf: [273,373], ht: 12	Reaction Vessel
1	fict_A Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
2	fict_B Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
3	fict_C Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
4	fict_D Vessel	pour by percent	fraction: [0,1]	Reaction Vessel

get_keys_to_action()

metadata: dict[str, Any] = {'render_fps': 60, 'render_modes': ['rgb_array']}

class chemistrylab.benches.reaction_bench.FictReact_v2

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action Index	Vessel	Event Performed	Event Parameters	Other Vessel
0	Reaction Vessel	heat contact	Tf: [273,373], ht: 300	Reaction Vessel
1	fict_A Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
2	fict_B Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
3	fict_C Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
4	fict_D Vessel	pour by percent	fraction: [0,1]	Reaction Vessel

metadata: dict[str, Any] = {'render_fps': 10, 'render_modes': ['rgb_array']}

class chemistrylab.benches.reaction_bench.GeneralWurtzReact_v0

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action Index	Vessel	Event Performed	Event Parameters	Other Vessel
0	Reaction Vessel	heat contact	Tf: [156,307], ht: 500	Reaction Vessel
1	1-chlorohexane Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
2	2-chlorohexane Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
3	3-chlorohexane Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
4	Na Vessel	pour by percent	fraction: [0,1]	Reaction Vessel

metadata: dict[str, Any] = {'render_fps': 10, 'render_modes': ['rgb_array']}

class chemistrylab.benches.reaction_bench.GeneralWurtzReact_v2

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action Index	Vessel	Event Performed	Event Parameters	Other Vessel
0	Reaction Vessel	heat contact	Tf: [156,307], ht: 500	Reaction Vessel
1	1-chlorohexane Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
2	2-chlorohexane Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
3	3-chlorohexane Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
4	Na Vessel	pour by percent	fraction: [0,1]	Reaction Vessel

metadata: dict[str, Any] = {'render_fps': 10, 'render_modes': ['rgb_array']}

class chemistrylab.benches.reaction_bench.WurtzReactDemo_v0

Bases: GenBench

Class to define an environment which performs a Wurtz extraction on materials in a vessel. Here is a breakdown of the action space:

Action Index	Vessel	Event Performed	Event Parameters	Other Vessel
0	Reaction Vessel	heat contact	Tf: [273,373], ht: 12	Reaction Vessel
1	1-chlorohexane Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
2	2-chlorohexane Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
3	3-chlorohexane Vessel	pour by percent	fraction: [0,1]	Reaction Vessel
4	Na Vessel	pour by percent	fraction: [0,1]	Reaction Vessel

get_keys_to_action()

metadata: dict[str, Any] = {'render_fps': 60, 'render_modes': ['rgb_array']}

chemistrylab.benches.reaction_bench.get_mat(mat, amount, name=None, volume=1.0)

Makes a Vessel with a single material

General Bench

class chemistrylab.benches.general_bench.Action(vessels, parameters, event_name, affected_vessels, dt, terminal)

Bases: NamedTuple

affected_vessels: Tuple[int] | None

Alias for field number 3

dt: float

Alias for field number 4

event_name: str

Alias for field number 2

parameters: Tuple[tuple]

Alias for field number 1

terminal: bool

Alias for field number 5

vessels: Tuple[int]

Alias for field number 0

class chemistrylab.benches.general_bench.ContinuousParam(min_val, max_val, thresh, other)

Bases: NamedTuple

max_val: float

Alias for field number 1

min_val: float

Alias for field number 0

other: tuple

Alias for field number 3

thresh: float

Alias for field number 2

class chemistrylab.benches.general_bench.GenBench(shelf: ~chemistrylab.lab.shelf.Shelf, action_list: ~typing.Tuple[~chemistrylab.benches.general_bench.Action], observation_list: ~typing.Tuple[str], targets: ~typing.Tuple[str], default_events: ~typing.Tuple[~chemistrylab.vessel.Event] = (), reward_function: ~typing.Callable = <function default_reward>, discrete=True, max_steps=50)

Bases: Env

A class representing an bench setup for conducting experiments.

This class represents a bench setup. The setup consists of a shelf with vessels, an action list that describes the actions that can be performed on the vessels by the agent, and a set of events which are performed on the vessels at the end of each step to update it’s state (ex. a chemical reaction).

Parameters:

• shelf (Shelf) – A shelf object to store vessels.

• action_list (list) – A list of Action objects containing information describing what each action is.

• observation_list (Tuple[str]) – List of observations to obtain from the characteriation bench

• targets (Tuple[str]) – The target materials for this bench

• default_events (Tuple[Event]) – A list of events to be performed on all working vessels in a bench at the end of each step.

• reward_function (Callable) – A function which accepts a target and a list of vessels and outputs a reward.

• discrete (bool) – set to True for a discrete action space and False for a continuous one

• max_steps (int) – Maximum number of steps for an episode

build_event(action, param)

Builds a list of (index, Event) tuples where index is the index of the vessel the Event will occur in.

Parameters:

• action (Action) – An action containing information about the vessels and parameters involved in the event.

• param (tuple) – A tuple containing parameters for the event.

Returns:

A list of (index, Event) tuples, where index is the index of the vessel the Event will occur in.

Return type:

List[Tuple]

get_target()

re_init()

Lets the bench know the current shelf setup is the starting setup. (i.e Resets the steps, rebuilds the list of actions, and resets the initial reward)

render()

Compute the render frames as specified by render_mode during the initialization of the environment.

The environment’s metadata render modes (env.metadata[“render_modes”]) should contain the possible ways to implement the render modes. In addition, list versions for most render modes is achieved through gymnasium.make which automatically applies a wrapper to collect rendered frames.

Note

As the render_mode is known during __init__, the objects used to render the environment state should be initialised in __init__.

By convention, if the render_mode is:

• None (default): no render is computed.

• “human”: The environment is continuously rendered in the current display or terminal, usually for human consumption. This rendering should occur during step() and render() doesn’t need to be called. Returns None.

• “rgb_array”: Return a single frame representing the current state of the environment. A frame is a np.ndarray with shape (x, y, 3) representing RGB values for an x-by-y pixel image.

• “ansi”: Return a strings (str) or StringIO.StringIO containing a terminal-style text representation for each time step. The text can include newlines and ANSI escape sequences (e.g. for colors).

• “rgb_array_list” and “ansi_list”: List based version of render modes are possible (except Human) through the wrapper, gymnasium.wrappers.RenderCollection that is automatically applied during gymnasium.make(..., render_mode="rgb_array_list"). The frames collected are popped after render() is called or reset().

Note

Make sure that your class’s metadata "render_modes" key includes the list of supported modes.

Changed in version 0.25.0: The render function was changed to no longer accept parameters, rather these parameters should be specified in the environment initialised, i.e., gymnasium.make("CartPole-v1", render_mode="human")

reset(*args, seed=None, options=None)

Resets the environment to an initial internal state, returning an initial observation and info.

This method generates a new starting state often with some randomness to ensure that the agent explores the state space and learns a generalised policy about the environment. This randomness can be controlled with the seed parameter otherwise if the environment already has a random number generator and reset() is called with seed=None, the RNG is not reset.

Therefore, reset() should (in the typical use case) be called with a seed right after initialization and then never again.

For Custom environments, the first line of reset() should be super().reset(seed=seed) which implements the seeding correctly.

Changed in version v0.25: The return_info parameter was removed and now info is expected to be returned.

Parameters:

• seed (optional int) – The seed that is used to initialize the environment’s PRNG (np_random). If the environment does not already have a PRNG and seed=None (the default option) is passed, a seed will be chosen from some source of entropy (e.g. timestamp or /dev/urandom). However, if the environment already has a PRNG and seed=None is passed, the PRNG will not be reset. If you pass an integer, the PRNG will be reset even if it already exists. Usually, you want to pass an integer right after the environment has been initialized and then never again. Please refer to the minimal example above to see this paradigm in action.

• options (optional dict) – Additional information to specify how the environment is reset (optional, depending on the specific environment)

Returns:

Observation of the initial state. This will be an element of observation_space

(typically a numpy array) and is analogous to the observation returned by step().

info (dictionary): This dictionary contains auxiliary information complementing observation. It should be analogous to

the info returned by step().

Return type:

observation (ObsType)

reset_with_target(target: str)

Resets the bench with a specific target material.

Parameters:

target (str) – The target material

set_target(target)

step(action)

Here, actions are performed by ‘_perform_discrete_action’ or ‘_perform_continuous_action’. Afterwards, all vessels in react_list have their concentrations updated by a reaction (specified in __init__). Finally, a CharacterizationBench is used to generate an observation, and a reward function provides a reward (if a terminal state was reached).

Parameters:

action (int or 1D array) – The action to be performed

validate_shelf()

Returns:

True if there are enough vessels to perform all of the actions and false otherwise.

Return type:

bool

chemistrylab.benches.general_bench.default_reward(vessels, targ)

Module contents