anatools.anaclient.ml module

Machine Learning Functions

create_ml_inference(self, datasetId, modelId, mapId=None, tags=None, workspaceId=None)

Create a new machine learning inference job.

Parameters

  • datasetId (str) – Dataset ID

  • modelId (str) – Model ID

  • mapId (str) – Map ID

  • workspaceId (str) – Workspace ID

Returns

Inference ID

Return type

str

create_ml_model(self, datasetId, architectureId, name, parameters, description=None, tags=None, workspaceId=None)

Creates a new machine learning model.

Parameters

  • architectureId (str) – Architecture ID

  • datasetId (str) – Dataset ID

  • name (str) – Model name

  • description (str) – Model description

  • paramters (str) – JSON string of model parameters

  • workspaceId (str) – Workspace ID

Returns

Machine learning model ID

Return type

str

delete_ml_inference(self, inferenceId, workspaceId=None)

Deletes or cancels a machine learning inference job.

Parameters

  • inferenceId (str) – Inference ID

  • workspaceId (str) – Workspace ID

Returns

Returns True if successful

Return type

bool

delete_ml_model(self, modelId, workspaceId=None)

Deletes or cancels a machine learning training job.

Parameters

  • modelId (str) – Model ID

  • workspaceId (str) – Workspace ID

Returns

Returns True if successful

Return type

bool

download_ml_inference(self, inferenceId, localDir=None, workspaceId=None)

Download the inference detections.

Parameters

  • inferencId (str) – Inference ID

  • localDir (str) – Local directory to save the model

  • workspaceId (str) – Workspace ID

Returns

Returns the filename of the downloaded model

Return type

str

download_ml_model(self, modelId, checkpoint=None, localDir=None, workspaceId=None)

Download the machine learning model.

Parameters

  • modelId (str) – Model ID

  • checkpoint (str) – Checkpoint to download. If not specified, the final model will be downloaded

  • localDir (str) – Local directory to save the model

  • workspaceId (str) – Workspace ID

Returns

Returns the filename of the downloaded model

Return type

str

edit_ml_inference(self, inferenceId, tags=None, workspaceId=None)

Edit the tags of a machine learning inference job.

Parameters

  • inferenceId (str) – Inference ID

  • tags (list[str]) – Tags to add or remove

  • workspaceId (str) – Workspace ID

Returns

Returns True if successful

Return type

bool

edit_ml_model(self, modelId, name=None, description=None, tags=None, workspaceId=None)

Edit the name or description of a machine learning model.

Parameters

  • modelId (str) – Model ID

  • name (str) – Model name

  • description (str) – Model description

  • tags (list[str]) – Model tags

  • workspaceId (str) – Workspace ID

Returns

Returns True if successful

Return type

bool

get_ml_architectures(self, fields=None)

Retrieves the machine learning model architectures available on the platform.

Parameters

fields (list[str], optional) – The fields to retrieve from the response.

Returns

Machine learning model architectures

Return type

dict

get_ml_inference_metrics(self, inferenceId, workspaceId=None)

Get the metrics from an inference job.

Parameters

  • inferenceId (str) – Inference ID

  • workspaceId (str) – Workspace ID

Returns

Metric data

Return type

dict

get_ml_inferences(self, workspaceId=None, inferenceId=None, datasetId=None, modelId=None, cursor=None, limit=None, filters=None, fields=None)

Get the inferences of a machine learning model.

Parameters

  • inferenceId (str) – Inference ID

  • datasetId (str) – Dataset ID

  • modelId (str) – Model ID

  • workspaceId (str) – Workspace ID

  • cursor (str) – Cursor for pagination

  • limit (int) – Maximum number of inferences to return

  • filters (dict) – Filters that limit output to entries that match the filter

  • fields (list) – List of fields to return, leave empty to get all fields.

Returns

Inference data

Return type

dict

get_ml_models(self, workspaceId=None, datasetId=None, modelId=None, cursor=None, limit=100, filters=None, fields=None)

Retrieves the machine learning model architectures available on the platform.

Parameters

  • workspaceId (str) – Workspace ID

  • datasetId (str) – Dataset ID

  • modelId (str) – Model ID

  • cursor (str, optional) – Cursor for pagination

  • limit (int, optional) – Maximum number of ml models to return

  • filters (dict) – Filters that limit output to entries that match the filter

  • fields (list[str], optional) – The fields to retrieve from the response.

Returns

Machine learning model architectures

Return type

dict

upload_ml_model(self, name, modelfile, architectureId, description=None, tags=None, workspaceId=None)

Upload a machine learning model.

Parameters

  • name (str) – Model name

  • modelfile (str) – The filepath of the compressed file containing the model, classes and spec.

  • architectureId (str) – Architecture ID

  • description (str) – Model description

  • tags (list[str]) – Model tags

  • workspaceId (str) – Workspace ID

Returns

Success / failure

Return type

bool