# emdgroup/baybe architecture

> Architecture diagram of the GitHub repository emdgroup/baybe, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-06). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/emdgroup/baybe
- Repository: https://github.com/emdgroup/baybe

## Overview

BayBE is a Bayesian design-of-experiments library: users define parameters, a constrained search space, and an objective, then submit measurements and request candidate configurations in a repeatable campaign loop. Campaign orchestration connects these domain objects to recommendation strategies; Bayesian strategies use surrogate models and acquisition functions. The README also documents bandit optimization, multi-target objectives, transfer learning, serialization, insights, and simulation/backtesting. Edges reflect the README and SOURCE INDEX references; unsampled feature implementations are represented at their supported directory paths, without asserting unverified internal calls.

Main entry points: [`baybe/campaign.py`](https://github.com/emdgroup/baybe/blob/main/baybe/campaign.py) (Campaign), [`baybe/recommenders/pure/bayesian/botorch`](https://github.com/emdgroup/baybe/tree/main/baybe/recommenders/pure/bayesian/botorch) (Bayesian recommenders), [`baybe/searchspace/core.py`](https://github.com/emdgroup/baybe/blob/main/baybe/searchspace/core.py) (Search space), [`baybe/recommenders/meta/sequential.py`](https://github.com/emdgroup/baybe/blob/main/baybe/recommenders/meta/sequential.py) (Sequential strategy)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_campaign["Campaign workflow"]
  node_campaign_api["Campaign<br/>[campaign.py]"]
  node_measurement["Measured outcomes"]
end

subgraph group_domain["Problem definition"]
  node_parameters["Parameters<br/>[base.py]"]
  node_searchspace["Search space<br/>[core.py]"]
  node_constraints["Constraints<br/>[base.py]"]
  node_targets["Targets<br/>[numerical.py]"]
  node_objectives["Objectives<br/>[base.py]"]
  node_transformations["Target transforms<br/>[basic.py]"]
end

subgraph group_recommendation["Recommendation engine"]
  node_recommender["Recommenders<br/>[base.py]"]
  node_sequential["Sequential strategy<br/>[sequential.py]"]
  node_naive["Bandit and naive<br/>[naive.py]"]
  node_bayesian["Bayesian recommenders"]
  node_acquisition["Acquisition functions<br/>[acqfs.py]"]
  node_surrogate["Surrogates<br/>[base.py]"]
  node_gp["Gaussian processes<br/>[core.py]"]
end

subgraph group_evaluation["Evaluation and operations"]
  node_simulation["Simulation<br/>[core.py]"]
  node_transfer["Transfer learning"]
  node_insights["Campaign insights<br/>[shap.py]"]
  node_serialization["Serialization<br/>[mixin.py]"]
end

node_user(("Experimenter"))

node_user -->|"configures and requests"| node_campaign_api
node_campaign_api -->|"uses"| node_searchspace
node_campaign_api -->|"uses"| node_objectives
node_campaign_api -->|"dispatches recommendations"| node_recommender
node_parameters -->|"builds subspaces"| node_searchspace
node_searchspace -->|"validates constraints"| node_constraints
node_objectives -->|"models targets"| node_targets
node_targets -->|"transforms values"| node_transformations
node_campaign_api -->|"accepts measurements"| node_measurement
node_measurement -->|"informs next cycle"| node_campaign_api
node_sequential -->|"selects strategy"| node_bayesian
node_sequential -->|"selects strategy"| node_naive
node_bayesian -->|"fits models"| node_surrogate
node_bayesian -->|"optimizes acquisition"| node_acquisition
node_gp -->|"implements model"| node_surrogate
node_user -->|"benchmarks settings"| node_simulation
node_simulation -->|"runs campaign"| node_campaign_api
node_user -.->|"supplies prior data"| node_transfer
node_user -->|"analyzes campaign"| node_insights
node_campaign_api -->|"supports persistence"| node_serialization
node_user -->|"stores and restores"| node_serialization

click node_campaign_api "https://github.com/emdgroup/baybe/blob/main/baybe/campaign.py"
click node_parameters "https://github.com/emdgroup/baybe/blob/main/baybe/parameters/base.py"
click node_searchspace "https://github.com/emdgroup/baybe/blob/main/baybe/searchspace/core.py"
click node_constraints "https://github.com/emdgroup/baybe/blob/main/baybe/constraints/base.py"
click node_targets "https://github.com/emdgroup/baybe/blob/main/baybe/targets/numerical.py"
click node_objectives "https://github.com/emdgroup/baybe/blob/main/baybe/objectives/base.py"
click node_transformations "https://github.com/emdgroup/baybe/blob/main/baybe/transformations/basic.py"
click node_recommender "https://github.com/emdgroup/baybe/blob/main/baybe/recommenders/base.py"
click node_sequential "https://github.com/emdgroup/baybe/blob/main/baybe/recommenders/meta/sequential.py"
click node_naive "https://github.com/emdgroup/baybe/blob/main/baybe/recommenders/naive.py"
click node_bayesian "https://github.com/emdgroup/baybe/tree/main/baybe/recommenders/pure/bayesian/botorch"
click node_acquisition "https://github.com/emdgroup/baybe/blob/main/baybe/acquisition/acqfs.py"
click node_surrogate "https://github.com/emdgroup/baybe/blob/main/baybe/surrogates/base.py"
click node_gp "https://github.com/emdgroup/baybe/blob/main/baybe/surrogates/gaussian_process/core.py"
click node_simulation "https://github.com/emdgroup/baybe/blob/main/baybe/simulation/core.py"
click node_transfer "https://github.com/emdgroup/baybe/blob/main/baybe/simulation/transfer_learning.py"
click node_insights "https://github.com/emdgroup/baybe/blob/main/baybe/insights/shap.py"
click node_serialization "https://github.com/emdgroup/baybe/blob/main/baybe/serialization/mixin.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_campaign_api,node_measurement toneBlue
class node_parameters,node_searchspace,node_constraints,node_targets,node_objectives,node_transformations toneAmber
class node_recommender,node_sequential,node_naive,node_bayesian,node_acquisition,node_surrogate,node_gp toneMint
class node_simulation,node_transfer,node_insights,node_serialization toneRose
class node_user toneIndigo
```

## Components

### Campaign workflow

- [Campaign](https://github.com/emdgroup/baybe/blob/main/baybe/campaign.py): `baybe/campaign.py`
- **Measured outcomes**

### Problem definition

- [Parameters](https://github.com/emdgroup/baybe/blob/main/baybe/parameters/base.py): `baybe/parameters/base.py`
- [Search space](https://github.com/emdgroup/baybe/blob/main/baybe/searchspace/core.py): `baybe/searchspace/core.py`
- [Constraints](https://github.com/emdgroup/baybe/blob/main/baybe/constraints/base.py): `baybe/constraints/base.py`
- [Targets](https://github.com/emdgroup/baybe/blob/main/baybe/targets/numerical.py): `baybe/targets/numerical.py`
- [Objectives](https://github.com/emdgroup/baybe/blob/main/baybe/objectives/base.py): `baybe/objectives/base.py`
- [Target transforms](https://github.com/emdgroup/baybe/blob/main/baybe/transformations/basic.py): `baybe/transformations/basic.py`

### Recommendation engine

- [Recommenders](https://github.com/emdgroup/baybe/blob/main/baybe/recommenders/base.py): `baybe/recommenders/base.py`
- [Sequential strategy](https://github.com/emdgroup/baybe/blob/main/baybe/recommenders/meta/sequential.py): `baybe/recommenders/meta/sequential.py`
- [Bandit and naive](https://github.com/emdgroup/baybe/blob/main/baybe/recommenders/naive.py): `baybe/recommenders/naive.py`
- [Bayesian recommenders](https://github.com/emdgroup/baybe/tree/main/baybe/recommenders/pure/bayesian/botorch): `baybe/recommenders/pure/bayesian/botorch`
- [Acquisition functions](https://github.com/emdgroup/baybe/blob/main/baybe/acquisition/acqfs.py): `baybe/acquisition/acqfs.py`
- [Surrogates](https://github.com/emdgroup/baybe/blob/main/baybe/surrogates/base.py): `baybe/surrogates/base.py`
- [Gaussian processes](https://github.com/emdgroup/baybe/blob/main/baybe/surrogates/gaussian_process/core.py): `baybe/surrogates/gaussian_process/core.py`

### Evaluation and operations

- [Simulation](https://github.com/emdgroup/baybe/blob/main/baybe/simulation/core.py): `baybe/simulation/core.py`
- [Transfer learning](https://github.com/emdgroup/baybe/blob/main/baybe/simulation/transfer_learning.py): `baybe/simulation/transfer_learning.py`
- [Campaign insights](https://github.com/emdgroup/baybe/blob/main/baybe/insights/shap.py): `baybe/insights/shap.py`
- [Serialization](https://github.com/emdgroup/baybe/blob/main/baybe/serialization/mixin.py): `baybe/serialization/mixin.py`

### Other components

- **Experimenter**

## Connections

- Experimenter → Campaign: configures and requests
- Campaign → Search space: uses (evidence: `baybe/campaign.py`)
- Campaign → Objectives: uses (evidence: `baybe/campaign.py`)
- Campaign → Recommenders: dispatches recommendations (evidence: `baybe/campaign.py`)
- Parameters → Search space: builds subspaces (evidence: `baybe/parameters/base.py`)
- Search space → Constraints: validates constraints (evidence: `baybe/searchspace/core.py`)
- Objectives → Targets: models targets (evidence: `baybe/objectives/base.py`)
- Targets → Target transforms: transforms values (evidence: `baybe/targets/numerical.py`)
- Campaign → Measured outcomes: accepts measurements
- Measured outcomes → Campaign: informs next cycle
- Sequential strategy → Bayesian recommenders: selects strategy (evidence: `baybe/recommenders/meta/sequential.py`)
- Sequential strategy → Bandit and naive: selects strategy (evidence: `baybe/recommenders/meta/sequential.py`)
- Bayesian recommenders → Surrogates: fits models
- Bayesian recommenders → Acquisition functions: optimizes acquisition (evidence: `baybe/acquisition/acqfs.py`)
- Gaussian processes → Surrogates: implements model (evidence: `baybe/surrogates/gaussian_process/core.py`)
- Experimenter → Simulation: benchmarks settings
- Simulation → Campaign: runs campaign (evidence: `baybe/simulation/scenarios.py`)
- Experimenter → Transfer learning: supplies prior data
- Experimenter → Campaign insights: analyzes campaign
- Campaign → Serialization: supports persistence (evidence: `baybe/campaign.py`)
- Experimenter → Serialization: stores and restores
