Artifact operations
MLFlowClient.listartifacts — Function
listartifacts(instance::MLFlow, run_id::String; path::String="", page_token::String="")
listartifacts(instance::MLFlow, run::Run; path::String="", page_token::String="")List artifacts for a Run.
Arguments
instance:MLFlowconfiguration.run_id: ID of theRunwhose artifacts to list.path: Filter artifacts matching this path (a relative path from the root artifact directory).page_token: Token indicating the page of artifact results to fetch
Returns
- Root artifact directory for the
Run. - List of file location and metadata for artifacts.
- Token that can be used to retrieve the next page of artifact results.
MLFlowClient.listartifactsdirect — Function
listartifactsdirect(instance::MLFlow; path::String="")List artifacts directly from the MLflow artifact repository (not associated with a run).
Arguments
instance:MLFlowconfiguration.path: Filter artifacts matching this path (a relative path from the root artifact directory).
Returns
List of file location and metadata for artifacts.
MLFlowClient.downloadartifact — Function
downloadartifact(instance::MLFlow, artifact_path::String)Download an artifact from the MLflow artifact repository.
Arguments
instance:MLFlowconfiguration.artifact_path: Path of the artifact to download.
Returns
Artifact binary data as a byte array.
MLFlowClient.uploadartifact — Function
uploadartifact(instance::MLFlow, artifact_path::String, data::Vector{UInt8})Upload an artifact to the MLflow artifact repository.
Arguments
instance:MLFlowconfiguration.artifact_path: Path where the artifact should be stored.data: Artifact binary data.
Returns
true if successful. Otherwise, raises exception.
MLFlowClient.deleteartifact — Function
deleteartifact(instance::MLFlow, artifact_path::String)Delete an artifact from the MLflow artifact repository.
Arguments
instance:MLFlowconfiguration.artifact_path: Path of the artifact to delete.
Returns
true if successful. Otherwise, raises exception.
MLFlowClient.createmultipartupload — Function
createmultipartupload(instance::MLFlow, artifact_path::String, num_parts::Int64)Create a multipart upload for an artifact.
Arguments
instance:MLFlowconfiguration.artifact_path: Path where the artifact should be stored.num_parts: Number of parts for the multipart upload.
Returns
upload_id: Upload ID for the multipart upload.credentials: Array of credentials for each part.
MLFlowClient.completemultipartupload — Function
completemultipartupload(instance::MLFlow, artifact_path::String, upload_id::String,
parts::Array{MultipartUploadPart})Complete a multipart upload.
Arguments
instance:MLFlowconfiguration.artifact_path: Path of the artifact.upload_id: Upload ID for the multipart upload.parts: Array of uploaded parts with their part numbers and ETags.
Returns
true if successful. Otherwise, raises exception.
MLFlowClient.abortmultipartupload — Function
abortmultipartupload(instance::MLFlow, artifact_path::String, upload_id::String)Abort a multipart upload.
Arguments
instance:MLFlowconfiguration.artifact_path: Path of the artifact.upload_id: Upload ID for the multipart upload.
Returns
true if successful. Otherwise, raises exception.
MLFlowClient.getpresigneddownloadurl — Function
getpresigneddownloadurl(instance::MLFlow, artifact_path::String)Get a presigned URL for downloading an artifact.
Arguments
instance:MLFlowconfiguration.artifact_path: Path of the artifact.
Returns
url: The presigned URL for downloading the artifact.headers: Optional headers that must be included in the download request.file_size: Optional size of the file in bytes.