ONNX Model Type

Upload an ONNX Model to a Machine Learning Repository

Use the following function to upload an ONNX model to a Machine Learning repository in Artifactory:

import frogml

repository = "repository-name"
name = "model-name"
namespace = "namespace"
version = "version-1"
properties = {"key1": "value1"}
dependencies = ["path/to/dependencies/pyproject.toml", "path/to/dependencies/poetry.lock"]
code_dir = "full/path/to/code/dir"
onnx_model = get_onnx_model() # Function that Returns a onnx model

frogml.onnx.log_model(
    model=onnx_model,
    repository=repository,
    model_name=name,
    namespace=namespace,
    version=version,
    properties=properties,
    dependencies=dependencies,
    code_dir=code_dir,
)

Download an ONNX Model from a Machine Learning Repository

Use the following function to download a deserialized ONNX model from a Machine Learning repository in Artifactory:

import frogml

repository = "repository-name"
name = "model-name"
namespace = "namespace"
version = "version-1"
full_target_path = "full/path/to/target/path"

onnx_deserialized_model = frogml.onnx.load_model(
    repository=repository,
    namespace=namespace,
    model_name=name,
    version=version,
    target_path=full_target_path,
)

Get Information on an ONNX Model in a Machine Learning Repository

Use the following function to retrieve information on a specific ONNX model stored in a Machine Learning repository in Artifactory.

import frogml

repository = "repository-name"
name = "model-name"
namespace = "namespace"
version = "version-1"

frogml.onnx.get_model_info(
    repository=repository,
    name=name,
    namespace=namespace,
    version=version,
)