GeomAIPredictions#

The Prediction module is in charge of running the GeomAI-powered predictions in your workspaces.

Directory#

class GeomAIPredictionDirectory#

Provides a collection of methods related to GeomAI model predictions.

This method is accessed through client.geomai.predictions.

Example

import ansys.simai.core as asc

simai_client = asc.from_config()
simai_client.geomai.predictions.list()
delete(prediction: GeomAIPrediction | str) → None#

Delete a specific GeomAI prediction from the server.

Parameters:

prediction (GeomAIPrediction | str) – ID or model of the prediction.

Raises:

NotFoundError – No prediction with the given ID exists.

download(prediction: GeomAIPrediction | str, file: BinaryIO | RawIOBase | BufferedIOBase | Path | str | PathLike | None = None) → None | BinaryIO#

Download the file generated by the prediction.

The downloaded file will be in the VTP format.

Parameters:
Returns:

None if a file is specified or a binary file-object otherwise.

Return type:

None | BinaryIO

Note

Consider using GeomAIPrediction.download() instead which will wait for the

prediction to be ready instead of throwing an error if it’s not.

get(id: str) → GeomAIPrediction#

Get a specific GeomAI prediction object from the server by ID.

Parameters:

id (str) – ID of the prediction.

Returns:

GeomAIPrediction instance with the given ID if it exists.

Raises:

NotFoundError – No prediction with the given ID exists.

Return type:

GeomAIPrediction

list(workspace: GeomAIWorkspace | str | None = None) → List[GeomAIPrediction]#

List all GeomAI predictions on the server that belong to the specified workspace or the configured one.

Parameters:

workspace (GeomAIWorkspace | str | None) – ID or model of the workspace to list the predictions for. This parameter is necessary if no workspace is set for the client.

run(configuration: GeomAIPredictionConfiguration | dict[str, Any], workspace: GeomAIWorkspace | str | None = None) → GeomAIPrediction#

Run a prediction in the given workspace with the given configuration.

Parameters:
Returns:

Created prediction object.

Raises:

ProcessingError – If the server failed to process the request.

Return type:

GeomAIPrediction

Examples

import ansys.simai.core as asc

simai_client = asc.from_config()
workspace = simai_client.geomai.workspaces.list()[0]
prediction = simai_client.geomai.predictions.run(
    dict(latent_params=[0.1, 1.2, 0.76], resolution=(100, 100, 100)),
    workspace,
)
sample(resolution: Tuple[Annotated[int, Gt(gt=0)], Annotated[int, Gt(gt=0)], Annotated[int, Gt(gt=0)]] | None = None, workspace: GeomAIWorkspace | str | None = None) → GeomAIPrediction#

Generate a sample GeomAI prediction object, in the given workspace with the given resolution.

Warning

This feature is in beta. It will likely be recast or retired by July 2026. Testing and feedback are encouraged, but it is advised not to include it in a production workflow.

Parameters:
Returns:

Created prediction object.

Return type:

GeomAIPrediction

Model#

class GeomAIPrediction#

Provides the local representation of a GeomAI prediction object.

delete() → None#

Remove a prediction from the server.

download(file: BinaryIO | RawIOBase | BufferedIOBase | Path | str | PathLike | None = None) → None | BinaryIO#

Download the file generated by the prediction.

The downloaded file will be in the VTP format.

Parameters:

file (BinaryIO | RawIOBase | BufferedIOBase | Path | str | PathLike | None) – Binary file-object or the path of the file to put the content into.

Returns:

None if a file is specified or a binary file-object otherwise.

Return type:

None | BinaryIO

reload() → None#

Refresh the object with its representation from the server.

wait(timeout: float | None = None) → bool#

Wait for all jobs concerning the object to either finish or fail.

Parameters:

timeout (float | None) – Maximum amount of time in seconds to wait. The default is None, it means that there is no maximum on the time to wait.

Returns:

True if the computation has finished, False if the operation timed out.

Return type:

bool

property configuration: GeomAIPredictionConfiguration#

The configuration used to run the prediction.

property failure_reason#

Optional message giving the causes for why the creation of the object failed.

See also

property fields: dict#

Dictionary containing the raw object representation.

property has_failed#

Boolean indicating if the creation of the object failed.

See also

property id: str#

ID of the object on the server.

property is_pending#

Boolean indicating if the object is still in creation. The value becomes False once object creation is either successful or has failed.

See also

property is_ready#

Boolean indicating if the object has finished creating without error.

See also

Configuration#

pydantic model GeomAIPredictionConfiguration#

The configuration used to run a GeomAI prediction.

Raises InvalidArguments if the input data cannot be validated to from a valid model.

field latent_params: List[float] [Required]#

A list of floats that represent the position of the geometry in the latent space.

These parameters describe the shape in a compressed form. The number of floats should match the nb_latent_param your model was requested with.

Required.

field resolution: Tuple[Annotated[int, Gt(gt=0)], Annotated[int, Gt(gt=0)], Annotated[int, Gt(gt=0)]] | None = None#

A list of three integers defining the number of voxels along the X, Y, and Z axes.

Use higher resolution for complex or precise geometries, and lower resolution for simple shapes or quick previews.

The total number of voxels must not exceed 900^3, that is x, y, z multiplied together must be less than or equal to 900^3. If you exceed that value, an error will occur.

Defaults to [100,100,100], if None is provided.

For the maximum resolution of 900^3, the prediction takes approximately 10 minutes (approximately 1 microsecond per voxel).