提供模型时 MLFlow 抛出错误：配置的跟踪 uri 方案：“文件”对于与代理 mlflow-artifact 方案一起使用无效

Question

我正在尝试使用 MLFlow CLI 在本地提供模型作为 REST 端点。 MLflow版本是2.11.3。以下是我正在使用的论点。

mlflow models serve -m "runs:/7782c4a700cc49c1a04f9ce608d90358/knnmodel" --env-manager local --port 5000

以下是例外情况：

Traceback (most recent call last):
  File "/home/vagrant/.local/bin/mlflow", line 8, in <module>
    sys.exit(cli())
  File "/home/vagrant/.local/lib/python3.8/site-packages/click/core.py", line 1157, in __call__
    return self.main(*args, **kwargs)
  File "/home/vagrant/.local/lib/python3.8/site-packages/click/core.py", line 1078, in main
    rv = self.invoke(ctx)
  File "/home/vagrant/.local/lib/python3.8/site-packages/click/core.py", line 1688, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "/home/vagrant/.local/lib/python3.8/site-packages/click/core.py", line 1688, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "/home/vagrant/.local/lib/python3.8/site-packages/click/core.py", line 1434, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/home/vagrant/.local/lib/python3.8/site-packages/click/core.py", line 783, in invoke
    return __callback(*args, **kwargs)
  File "/home/vagrant/.local/lib/python3.8/site-packages/mlflow/models/cli.py", line 104, in serve
    return get_flavor_backend(
  File "/home/vagrant/.local/lib/python3.8/site-packages/mlflow/models/flavor_backend_registry.py", line 44, in get_flavor_backend
    local_path = _download_artifact_from_uri(
  File "/home/vagrant/.local/lib/python3.8/site-packages/mlflow/tracking/artifact_utils.py", line 105, in _download_artifact_from_uri
    return get_artifact_repository(artifact_uri=root_uri).download_artifacts(
  File "/home/vagrant/.local/lib/python3.8/site-packages/mlflow/store/artifact/artifact_repository_registry.py", line 124, in get_artifact_repository
    return _artifact_repository_registry.get_artifact_repository(artifact_uri)
  File "/home/vagrant/.local/lib/python3.8/site-packages/mlflow/store/artifact/artifact_repository_registry.py", line 77, in get_artifact_repository
    return repository(artifact_uri)
  File "/home/vagrant/.local/lib/python3.8/site-packages/mlflow/store/artifact/runs_artifact_repo.py", line 27, in __init__
    self.repo = get_artifact_repository(uri)
  File "/home/vagrant/.local/lib/python3.8/site-packages/mlflow/store/artifact/artifact_repository_registry.py", line 124, in get_artifact_repository
    return _artifact_repository_registry.get_artifact_repository(artifact_uri)
  File "/home/vagrant/.local/lib/python3.8/site-packages/mlflow/store/artifact/artifact_repository_registry.py", line 77, in get_artifact_repository
    return repository(artifact_uri)
  File "/home/vagrant/.local/lib/python3.8/site-packages/mlflow/store/artifact/mlflow_artifacts_repo.py", line 45, in __init__
    super().__init__(self.resolve_uri(artifact_uri, get_tracking_uri()))
  File "/home/vagrant/.local/lib/python3.8/site-packages/mlflow/store/artifact/mlflow_artifacts_repo.py", line 59, in resolve_uri
    _validate_uri_scheme(track_parse.scheme)
  File "/home/vagrant/.local/lib/python3.8/site-packages/mlflow/store/artifact/mlflow_artifacts_repo.py", line 35, in _validate_uri_scheme
    raise MlflowException(
mlflow.exceptions.MlflowException: The configured tracking uri scheme: 'file' is invalid for use with the proxy mlflow-artifact scheme. The allowed tracking schemes are: {'https', 'http'}

似乎无法从 MLflow Web 服务器访问 MLflow 工件。有解决这个问题吗？

Answer 1

你需要导出这个变量

export MLFLOW_TRACKING_URI=http://127.0.0.1:<PORT of the mlflow server>