# Example for using the multi target mode for the objective Example for using the multi target mode for the objective. It uses a desirability value to handle several targets. This example assumes some basic familiarity with using BayBE. We thus refer to [`campaign`](./../Basics/campaign.md) for a basic example. ## Necessary imports for this example ```python import pandas as pd ``` ```python from baybe import Campaign from baybe.objectives import DesirabilityObjective from baybe.parameters import CategoricalParameter, NumericalDiscreteParameter from baybe.searchspace import SearchSpace from baybe.targets import NumericalTarget from baybe.utils.dataframe import add_fake_results ``` ## Experiment setup and creating the searchspace ```python Categorical_1 = CategoricalParameter("Cat_1", values=["22", "33"], encoding="OHE") Categorical_2 = CategoricalParameter( "Cat_2", values=["very bad", "bad", "OK", "good", "very good"], encoding="INT", ) Num_disc_1 = NumericalDiscreteParameter( "Num_disc_1", values=[1, 2, 3, 4, 6, 8, 10], tolerance=0.3 ) Num_disc_2 = NumericalDiscreteParameter( "Num_disc_2", values=[-1, -3, -6, -9], tolerance=0.3 ) ``` ```python parameters = [Categorical_1, Categorical_2, Num_disc_1, Num_disc_2] ``` ```python searchspace = SearchSpace.from_product(parameters=parameters) ``` ## Defining the targets The multi target mode is handled when creating the objective object. Thus we first need to define the different targets. This examples has different targets with different modes. The first target is maximized and while the second one is minimized. Note that in this multi target mode, the user must specify bounds for each target. ```python Target_1 = NumericalTarget( name="Target_1", mode="MAX", bounds=(0, 100), transformation="LINEAR" ) Target_2 = NumericalTarget( name="Target_2", mode="MIN", bounds=(0, 100), transformation="LINEAR" ) ``` For each target it is also possible to specify a `target_transform` function. A detailed discussion of this functionality can be found at the end of this example. In this example, define a third target working with the mode `MATCH`. We furthermore use `target_transform="BELL"`. ```python Target_3 = NumericalTarget( name="Target_3", mode="MATCH", bounds=(45, 55), transformation="BELL" ) ``` Note that the `MATCH` mode seeks to have the target at the mean between the two bounds. For example, choosing 95 and 105 will lead the algorithm seeking 100 as the optimal value. Thus, using the bounds, it is possible to control both the match target and the range around this target that is considered viable. ## Creating the objective Now to work with these three targets the objective object must be properly created. The mode is set to `DESIRABILITY` and the targets are described in a list. ```python targets = [Target_1, Target_2, Target_3] ``` As the recommender requires a single function, the different targets need to be combined. Thus, a `scalarizer` is used to create a single target out of the several targets given. The combine function can either be the mean `MEAN` or the geometric mean `GEOM_MEAN`. Per default, `GEOM_MEAN` is used. Weights for each target can also be specified as a list of floats in the arguments Per default, weights are equally distributed between all targets and are normalized internally. It is thus not necessary to handle normalization or scaling. ```python objective = DesirabilityObjective( targets=targets, weights=[20, 20, 60], scalarizer="MEAN", ) ``` ```python print(objective) ``` Objective Type: DesirabilityObjective Targets Type Name ... Transformation Weight 0 NumericalTarget Target_1 ... LINEAR 20.0 1 NumericalTarget Target_2 ... LINEAR 20.0 2 NumericalTarget Target_3 ... BELL 60.0 [3 rows x 7 columns] Scalarizer: MEAN ## Creating and printing the campaign ```python campaign = Campaign(searchspace=searchspace, objective=objective) print(campaign) ``` Campaign Meta Data Batches done: 0 Fits done: 0 SearchSpace Search Space Type: DISCRETE SubspaceDiscrete Discrete Parameters Name Type Num_Values Encoding 0 Cat_1 CategoricalP... 2 CategoricalE... 1 Cat_2 CategoricalP... 5 CategoricalE... 2 Num_disc_1 NumericalDis... 7 None 3 Num_disc_2 NumericalDis... 4 None Experimental Representation Cat_1 Cat_2 Num_disc_1 Num_disc_2 0 22 OK 1.0 -9.0 1 22 OK 1.0 -6.0 2 22 OK 1.0 -3.0 .. ... ... ... ... 277 33 very good 10.0 -6.0 278 33 very good 10.0 -3.0 279 33 very good 10.0 -1.0 [280 rows x 4 columns] Meta Data was_recommended: 0/280 was_measured: 0/280 dont_recommend: 0/280 Constraints Empty DataFrame Columns: [] Index: [] Computational Representation Cat_1_22 Cat_1_33 ... Num_disc_1 Num_disc_2 0 1.0 0.0 ... 1.0 -9.0 1 1.0 0.0 ... 1.0 -6.0 2 1.0 0.0 ... 1.0 -3.0 .. ... ... ... ... ... 277 0.0 1.0 ... 10.0 -6.0 278 0.0 1.0 ... 10.0 -3.0 279 0.0 1.0 ... 10.0 -1.0 [280 rows x 5 columns] Objective Type: DesirabilityObjective Targets Type Name ... Transformation Weight 0 NumericalTarget Target_1 ... LINEAR 20.0 1 NumericalTarget Target_2 ... LINEAR 20.0 2 NumericalTarget Target_3 ... BELL 60.0 [3 rows x 7 columns] Scalarizer: MEAN TwoPhaseMetaRecommender Initial recommender RandomRecommender Compatibility: SearchSpaceType.HYBRID Recommender BotorchRecommender Surrogate GaussianProcessSurrogate Supports Transfer Learning: True Kernel factory: DefaultKernelFactory() Acquisition function: qLogExpectedImprovement() Compatibility: SearchSpaceType.HYBRID Sequential continuous: False Hybrid sampler: None Sampling percentage: 1.0 Switch after: 1 ## Performing some iterations The following loop performs some recommendations and adds fake results. It also prints what happens to internal data. ```python N_ITERATIONS = 3 ``` ```python for kIter in range(N_ITERATIONS): rec = campaign.recommend(batch_size=3) add_fake_results(rec, campaign.targets) campaign.add_measurements(rec) desirability = campaign.objective.transform(campaign.measurements) print(f"\n\n#### ITERATION {kIter+1} ####") print("\nRecommended measurements with fake measured results:\n") print(rec) print("\nInternal measurement database with desirability values:\n") print(pd.concat([campaign.measurements, desirability], axis=1)) ``` #### ITERATION 1 #### Recommended measurements with fake measured results: Cat_1 Cat_2 Num_disc_1 Num_disc_2 Target_1 Target_2 Target_3 258 33 very bad 8.0 -3.0 71.043827 52.712361 55.997167 249 33 very bad 4.0 -6.0 99.827009 55.051652 54.884452 212 33 good 10.0 -9.0 35.124065 68.489852 55.844039 Internal measurement database with desirability values: Cat_1 Cat_2 Num_disc_1 Num_disc_2 Target_1 Target_2 Target_3 \ 0 33 very bad 8.0 -3.0 71.043827 52.712361 55.997167 1 33 very bad 4.0 -6.0 99.827009 55.051652 54.884452 2 33 good 10.0 -9.0 35.124065 68.489852 55.844039 BatchNr FitNr Desirability 0 1 NaN 0.528913 1 1 NaN 0.661878 2 1 NaN 0.436311 #### ITERATION 2 #### Recommended measurements with fake measured results: Cat_1 Cat_2 Num_disc_1 Num_disc_2 Target_1 Target_2 \ index 73 33 very bad 2.0 -6.0 67.467637 79.995753 253 33 very bad 6.0 -6.0 15.108640 30.687980 265 33 very good 4.0 -6.0 71.740403 63.771165 Target_3 index 73 54.493411 253 51.288282 265 54.988861 Internal measurement database with desirability values: Cat_1 Cat_2 Num_disc_1 Num_disc_2 Target_1 Target_2 Target_3 \ 0 33 very bad 8.0 -3.0 71.043827 52.712361 55.997167 1 33 very bad 4.0 -6.0 99.827009 55.051652 54.884452 2 33 good 10.0 -9.0 35.124065 68.489852 55.844039 3 33 very bad 2.0 -6.0 67.467637 79.995753 54.493411 4 33 very bad 6.0 -6.0 15.108640 30.687980 51.288282 5 33 very good 4.0 -6.0 71.740403 63.771165 54.988861 BatchNr FitNr Desirability 0 1 1.0 0.528913 1 1 1.0 0.661878 2 1 1.0 0.436311 3 2 NaN 0.575604 4 2 NaN 0.749252 5 2 NaN 0.580668 #### ITERATION 3 #### Recommended measurements with fake measured results: Cat_1 Cat_2 Num_disc_1 Num_disc_2 Target_1 Target_2 Target_3 index 213 33 good 10.0 -6.0 86.225597 65.220773 51.371454 261 33 very bad 10.0 -6.0 13.453378 85.487752 55.666635 145 33 OK 8.0 -6.0 47.268558 78.835603 53.284941 Internal measurement database with desirability values: Cat_1 Cat_2 Num_disc_1 Num_disc_2 Target_1 Target_2 Target_3 \ 0 33 very bad 8.0 -3.0 71.043827 52.712361 55.997167 1 33 very bad 4.0 -6.0 99.827009 55.051652 54.884452 2 33 good 10.0 -9.0 35.124065 68.489852 55.844039 3 33 very bad 2.0 -6.0 67.467637 79.995753 54.493411 4 33 very bad 6.0 -6.0 15.108640 30.687980 51.288282 5 33 very good 4.0 -6.0 71.740403 63.771165 54.988861 6 33 good 10.0 -6.0 86.225597 65.220773 51.371454 7 33 very bad 10.0 -6.0 13.453378 85.487752 55.666635 8 33 OK 8.0 -6.0 47.268558 78.835603 53.284941 BatchNr FitNr Desirability 0 1 1.0 0.528913 1 1 1.0 0.661878 2 1 1.0 0.436311 3 2 2.0 0.575604 4 2 2.0 0.749252 5 2 2.0 0.580668 6 3 NaN 0.819858 7 3 NaN 0.371607 8 3 NaN 0.620396 ## Addendum: Description of `transformation` functions This function is used to transform target values to the interval `[0,1]` for `MAX`/`MIN` mode. An ascending or decreasing `LINEAR` function is used per default. This function maps input values in a specified interval [lower, upper] to the interval `[0,1]`. Outside the specified interval, the function remains constant, that is, 0 or 1. For the match mode, two functions are available `TRIANGULAR` and `BELL`. The `TRIANGULAR` function is 0 outside a specified interval and linearly increases to 1 from both interval ends, reaching the value 1 at the center of the interval. This function is used per default for MATCH mode. The `BELL` function is a Gaussian bell curve, specified through the boundary values of the sigma interval, reaching the maximum value of 1 at the interval center.