module model.py


function calculate_model_size

Tells about the size of the model

Args:

  • model_dir (str): directory in which model is present.

Returns:
total size of the model


class ModelVersion

property created_at

Get the time at which model version was created


property created_by

Get the information about who created the model version


property description

Get description of the model


property fqn

Get fqn of the current model version


property metadata

Get metadata for the current model


property metrics

get the metrics for the current version of the model


property model_fqn

Get fqn of the model


property model_schema

get the model schema for current model


property name

Get the name of the model


property step

Get the step in which model was created


property updated_at

Get the information about when the model version was updated


property version

Get version information of the model


function delete

Deletes the current instance of the ModelVersion hence deleting the current version.

Returns:
True if model was deleted successfully

Examples:

import mlfoundry

client = mlfoundry.get_client()
model_version = client.get_model_version_by_fqn(fqn="<your-model-fqn>")
model_version.delete()
    

function download

Download an model file or directory to a local directory if applicable, and return a local path for it.

Args:

  • path (str): Absolute path of the local filesystem destination directory to which to download the specified models. This directory must already exist. If unspecified, the models will either be downloaded to a new uniquely-named directory on the local filesystem or will be returned directly in the case of the Local ModelRepository.
  • overwrite (bool): If True it will overwrite the file if it is already present in the download directory else it will throw an error

Returns:

  • path: Absolute path of the local filesystem location containing the desired models.

Examples:

import mlfoundry

client = mlfoundry.get_client()
model_version = client.get_model_version_by_fqn(fqn="<your-model-fqn>")
model_version.download(path="<your-desired-download-path>")
    

classmethod from_fqn

Get the version of a model to download contents or load them in memory

Args:

  • fqn (str): Fully qualified name of the model version.

Returns:

  • ModelVersion: An ModelVersion instance of the Model

Examples:

import mlfoundry

client = mlfoundry.get_client()
model_version = mlfoundry.ModelVersion.from_fqn(fqn="<your-model-fqn>")
    

function load


function raw_download

Download an model file or directory to a local directory if applicable, and return a local path for it.

Args:

  • path (str): Absolute path of the local filesystem destination directory to which to download the specified models. This directory must already exist. If unspecified, the models will either be downloaded to a new uniquely-named directory on the local filesystem or will be returned directly in the case of the Local ModelRepository.
  • overwrite (bool): If True it will overwrite the file if it is already present in the download directory else it will throw an error

Returns:

  • path: Absolute path of the local filesystem location containing the desired models.

Examples:

import mlfoundry

client = mlfoundry.get_client()
model_version = client.get_model_version_by_fqn(fqn="<your-model-fqn>")
model_version.raw_download(path="<your-desired-download-path>")
    

function update

Updates the current instance of the ModelVersion hence updating the current version.

Examples:

import mlfoundry

client = mlfoundry.get_client()
model_version = client.get_model_version_by_fqn(fqn="<your-model-fqn>")
model_version.description = 'This is the new description'
model_version.update()