Documentation

This is the documentation of the Grasshopper components in ARA.

The documentation of AIXD: AI-eXtended Design tookit can be found here.

DataBool

_images/ara_DataBool.png

Defines a boolean variable (True or False).

Inputs

  • name (str) – Name of the variable.

  • dim (int) – Dimension of the variable.

Outputs

  • dataobject – Data object.

DataCategorical

_images/ara_DataCat.png

Defines a categorical variable (for example, represening labels or classes).

Inputs

  • name (str) – Name of the variable.

  • dim (int) – Dimension of the variable.

  • options [List of (str)] – Options: list of possible categories, as strings.

Outputs

  • dataobject – Data object.

DataInt

_images/ara_DataInt.png

Defines an integer-valued variable.

Inputs

  • name (str) – Name of the variable.

  • dim (int) – Dimension of the variable.

  • domain (interval) – Domain of the variable as an interval.

Outputs

  • dataobject – Data object.

DataObjectsNames

_images/ara_DataObjectsNames.png

Generates panels with list of names of data objects for all existing data blocks.

Inputs

  • get_names (bool) – Set to True to run.

DataReal

_images/ara_DataReal.png

Defines a real-valued variable.

Inputs

  • name (str) – Name of the variable.

  • dim (int) – Dimension of the variable.

  • domain (interval) – Domain of the variable as an interval.

Outputs

  • dataobject – Data object.

DatasetCreate

_images/ara_DatasetCreate.png

Creates a dataset object. This defines the structure of the dataset. It does not cointain any data.

Inputs

  • design_parameters [List of (none)] – Design parameters: list of data objects.

  • performance_attributes [List of (none)] – Performance attributes: list of data objects.

  • create (bool) – Set to True to create a dataset object. If a dataset already exists in the project path, nothing will happen. To create a new dataset, change the project path or dataset name or delete the existing dataset.

Outputs

  • msg – Message or error.

DatasetGenerator

_images/ara_DatasetGenerator.png

Provides instructions on how to generate random samples for the dataset by harnessing the parametric model found in the current Grasshopper file. See Tutorial for more information on how to prepare the inputs and outputs of the parametric model.

Outputs

  • instructions – Information on how to run the dataset_generator script.

DatasetLoad

_images/ara_DatasetLoad.png

Loads an existing dataset from the file system, from the location specified in the project setup. It loads the dataset object and the data into the app.

Inputs

  • load (bool) – Set to True to load the dataset.

Outputs

  • msg –

DatasetOneSample

_images/ara_DatasetOneSample.png

Retrieves one sample from the dataset (at a given or random index) and instantiates it in the parametric model. Requires a dataset to be loaded.

Inputs

  • item (int) – Index of the sample in the dataset, optional. If not provided, a random index will be selected.

  • get (bool) – Set to True to retrieve a sample.

Outputs

  • sample_summary – Summary of the retrieved sample.

DatasetSummary

_images/ara_DatasetSummary.png

Provides a summary of the dataset. Requires a dataset to be loaded.

Inputs

  • get (bool) – Set to True to get the summary of the dataset.

Outputs

  • summary – Summary of the dataset.

Generator

_images/ara_Generator.png

Runs a generation campaing to create new designs using the trained model. Requires a dataset and a trained model to be loaded.

Inputs

  • requested_values [List of (str)] – List of requested values, each formatted as a string with the following format: ‘variable_name:value’.

  • n_designs (int) – Number of designs to generate.

  • generate (bool) – Set to True to start the generation process.

  • clear (bool) – Forget the previously generated designs.

  • pick_previous (bool) – Iterate backward through the list of generated designs, instantiate the previous sample.

  • pick_next (bool) – Iterate forward through the list of generated designs, instantiate the next sample.

Outputs

  • sample_summary – Selected sample.

ModelDimensions

_images/ara_ModelDims.png

Retrieves dimensions of the model’s input and output layers. Requires that a model has been set up or loaded.

Inputs

  • get (bool) – Set to True to retrieve input and output dimensions of the model.

Outputs

  • summary – Summary of the model’s input and output dimensions.

ModelLoad

_images/ara_ModelLoad.png

Loads an existing, pre-traind neural network model from a checkpoint. Requires a dataset to be loaded.

Inputs

  • model_type (str) – Type of the autoencoder model. Options are: ‘CAE’ (conditional Autoencoder) and ‘CVAE’ (conditional Variational Autoencoder). Default: ‘CAE’.

  • checkpoint_name (str) – Name of the checkpoint file to load the model from, without the file extension. The file’s extension must be .ckpt

  • checkpoint_path (str) – Path to the directory containing the checkpoint file.

  • load (bool) – Set to True to load the model.

Outputs

  • msg – Confirmation of the model loading, or an error message.

ModelSetup

_images/ara_ModelSetup.png

Sets up an autoencoder model of the specified type with the given parameters. Requires a dataset to be loaded.

Inputs

  • model_type (str) – Type of the autoencoder model. Options are: ‘CAE’ (conditional Autoencoder) and ‘CVAE’ (conditional Variational Autoencoder). Default: ‘CAE’.

  • features [List of (str)] – List of variable names to be used as input to the model.

  • targets [List of (str)] – List of variable names to be used as output from the model.

  • latent_dim (int) – Dimension of the latent space.

  • hidden_layers [List of (int)] – Width of each hidden layer (list of int).

  • batch_size (int) – Size of the training batches

  • set (bool) – Set to True to set up the model.

Outputs

  • quick_summary – Quick summary of the model.

  • model_dims – Input and output dimensions of the model.

ModelSummary

_images/ara_ModelSummary.png

Provides a summary of the autoencoder model’s architecture. Requires that a model has been set up or loaded.

Inputs

  • max_depth (int) – Sets the depth of the summary. The larger the depth, the more detailed the summary.

  • get (bool) – Retrieves the model information.

Outputs

  • summary – Model summary.

ModelTrain

_images/ara_ModelTrain.png

Runs a training campaign. Requires that a model has been set up (to train from scratch) or loaded (to continue training).

Inputs

  • epochs (int) – Number of training epochs.

  • wb (str) – Weights&Biases: username or team name. If not set, W&B will not be used.

  • run (bool) – Set to True to start training.

Outputs

  • best_ckpt – Filename of the best performing checkpoint.

  • path – Path to all checkpoints.

PlotContours

_images/ara_PlotContours.png

Plots the distribution contours for each pair of variables from the data in the dataset. Launches an interactive plot in a browser. Requires a dataset to be loaded.

Inputs

  • variables [List of (str)] – List of names of the variables to be plotted.

  • plot (bool) – Set to True to (re-)create the plot.

PlotContoursRequest

_images/ara_PlotContoursRequest.png

Plots the requested and predicted values against the distribution contours for each pair of the corresponding variables. Launches an interactive plot in a browser. Requires a trained model to be loaded.

Inputs

  • request [List of (str)] – List of requested values, each formatted as a string with the following format: ‘variable_name:value’.

  • plot (bool) – Set to True to (re-)create the plot.

PlotCorrelations

_images/ara_PlotCorrelations.png

Plots correlation matrix for the given variables from the data in the dataset. Launches an interactive plot in a browser. Requires a dataset to be loaded.

Inputs

  • variables [List of (str)] – List of names of the variables to be plotted.

  • plot (bool) – Set to True to (re-)create the plot.

PlotDistribution

_images/ara_PlotDistributions.png

Plots the distribution of the given variables from the data in the dataset. Launches an interactive plot in a browser. Requires a dataset to be loaded.

Inputs

  • variables [List of (str)] – List of names of the variables to be plotted.

  • plot (bool) – Set to True to (re-)create the plot.

ProjectSetup

_images/ara_ProjectSetup.png

Sets up the project in the folder given by project_root/project_name.

Inputs

  • set (bool) –

  • project_root (str) – Path to the project root folder. If none is given, the default is the parent folder of this Grasshopper file.

  • project_name (str) – Any name for the project. It will be used to create a folder with the same name in the project root folder. All files will be later saved here.

Outputs

  • msg – Messages and errors.

  • path – Effective path to the project.

Reset

_images/ara_Reset.png

Resets the current project running in this Grasshopper file.

Inputs

  • reset (bool) – Set to True to reset.

Server

_images/ara_Server.png

Starts and stops the app server.

Inputs

  • start (bool) – Starts the server.

  • stop (bool) – Stops the server.

  • show_window (bool) – If True, the server window will be shown. If False, the server window will be hidden. Default: True.

Outputs

  • msg – Messages or errors.

ShowFolder

_images/ara_ShowFolder.png

Reveals the folder in the file explorer.

Inputs

  • path (str) – Path to the (local) folder.

  • open (bool) – Set to True to open the folder in the file explorer.