Using ML Flow>train projects remote throwing error for conda env path

从 notebooks/how-to-use-azureml/track-and-monitor-experiments/using-mlflow/train-projects-remote/train-projects-remote.ipynb 和第 15 个单元格中失败的 mlflow.projects.run 调用开始。跟踪 Azure ML 后端对 Project 对象的处理以及其 conda 环境查找过程。当远程实验能够在不出现 conda_env_path AttributeError 的情况下运行时，即表示完成。

When executing the train-projects-remote notebook in VS Code it throws and error when you try to run the experiment.
AttributeError: 'Project' object has no attribute 'conda_env_path'

Output exceeds the size limit. Open the full output data in a text editor

AttributeError Traceback (most recent call last)
/home/brcampb/projects/ml/notebooks/how-to-use-azureml/track-and-monitor-experiments/using-mlflow/train-projects-remote/train-projects-remote.ipynb Cell 15 in <cell line: 1>()
----> 1 remote_mlflow_run = mlflow.projects.run(uri=".",
2 parameters={"alpha":0.3},
3 backend = "azureml",
4 backend_config = backend_config,
5 synchronous=True)

File ~/miniconda3/envs/azureml/lib/python3.8/site-packages/mlflow/projects/init.py:331, in run(uri, entry_point, version, parameters, docker_args, experiment_name, experiment_id, backend, backend_config, use_conda, storage_dir, synchronous, run_id, run_name, env_manager)
325 backend_config_dict[MLFLOW_LOCAL_BACKEND_RUN_ID_CONFIG] = run_id
327 experiment_id = _resolve_experiment_id(
328 experiment_name=experiment_name, experiment_id=experiment_id
329 )
--> 331 submitted_run_obj = _run(
332 uri=uri,
333 experiment_id=experiment_id,
334 entry_point=entry_point,
335 version=version,
336 parameters=parameters,
337 docker_args=docker_args,
338 backend_name=backend,
339 backend_config=backend_config_dict,
340 env_manager=env_manager,
341 storage_dir=storage_dir,
...
---> 96 if mlproject.conda_env_path:
97 _logger.info(_CONSOLE_MSG.format("Creating remote conda environment for project using MLproject"))
98 environment = Environment.from_conda_specification(name="environment", file_path=mlproject.conda_env_path)

AttributeError: 'Project' object has no attribute 'conda_env_path'