Updated install.md

docs/install.md

@@ -1,43 +1,80 @@
-# Installation
-
-In BoFire we have several optional dependencies.
-
-### Domain and Optimization Algorithms
-
-To install BoFire with optimization tools you can use
-```
-pip install bofire[optimization]
-```
-This will also install [BoTorch](https://botorch.org/) that depends on [PyTorch](https://pytorch.org/).
-### Design of Experiments
-
-BoFire has functionality to create D-optimal experimental designs via the `doe` module. This module is depends on
-[Cyipopt](https://cyipopt.readthedocs.io/en/stable/). A comfortable way to install Cyipopt and the dependencies is via
-```
-conda install -c conda-forge cyipopt
-```
-You have to install Cyipopt manually.
-### Just Domain
-
-If you just want a data structure that represents the domain of an optimization problem you can
-```
-pip install bofire
-```
-
-### Cheminformatics
-
-Some features related to molecules and their representation depend on [Rdkit](https://www.rdkit.org/).
-```
-pip install bofire[optimization,cheminfo]
-```
-
-### Development Installation
-If you want to [contribute](CONTRIBUTING.md) to BoFire, you might want to install in editable mode including the test dependencies.
-After cloning the repository via
-```
-git clone https://github.com/experimental-design/bofire.git
-```
-and cd `bofire`, you can proceed with
-```
-pip install -e .[optimization,cheminfo,docs,tests]
-```
+<h1><img src="..\graphics\logos\bofire-long.png" alt="BoFire" style="vertical-align: middle;"> Installation Guide</h1>
+
+### Installation from package index
+
+If you just want a data structure that represents the domain of an optimization problem you can run :
+
+```
+pip install bofire
+```
+
+### Additional optional dependencies
+
+In BoFire we have several optional dependencies.
+
+- `[optimization]`\: access to optimization tools
+- `[cheminfo]`\: includes features related to molecules and their representation
+- `[entmoot]`\: enable the ensemble tree model optimization tool
+- `[xgb]`\: for extreme gradient boosting
+- `[all]`\: install all possible options (except DoE)
+- `Design of Experiments (DoE)`\: manual installation necessary
+
+### Domain and Optimization Algorithms
+
+To install BoFire with optimization tools you can use
+
+```
+pip install bofire[optimization]
+```
+
+This will also install [BoTorch](https://botorch.org/) that depends on [PyTorch](https://pytorch.org/).
+
+### Cheminformatics
+
+Some features related to molecules and their representation depend on [Rdkit](https://www.rdkit.org/).
+
+```
+pip install bofire[cheminfo]
+```
+
+### Ensemble Tree Model
+
+This option enables tree-based optimization using the Ensemble Tree Model Optimization Tool using tree-based surrogate models and will install [entmoot](https://entmoot.readthedocs.io/) and [lightgbm](https://lightgbm.readthedocs.io/).
+
+```
+pip install bofire[entmoot]
+```
+
+### Extreme Gradient Boost
+
+For extreme gradient boosting this option will also install [xgboost](https://xgboost.readthedocs.io/).
+
+```
+pip install bofire[xgb]
+```
+
+### Design of Experiments
+
+BoFire has functionality to create D-optimal experimental designs via the `doe` module. This module is depends on
+[Cyipopt](https://cyipopt.readthedocs.io/en/stable/). A comfortable way to install Cyipopt and the dependencies is via
+
+```
+conda install -c conda-forge cyipopt
+```
+
+You have to install Cyipopt manually.
+
+### Development Installation
+
+If you want to [contribute](CONTRIBUTING.md) to BoFire, you might want to install in editable mode including the test dependencies.
+After cloning the repository via
+
+```
+git clone https://github.com/experimental-design/bofire.git
+```
+
+and cd `bofire`, you can proceed with
+
+```
+pip install -e .[optimization,cheminfo,docs,tests]
+```