# mlflow
pip install git+https://github.com/mlflow/mlflow.git@refs/pull/18671/merge
# mlflow-skinny
pip install git+https://github.com/mlflow/mlflow.git@refs/pull/18671/merge#subdirectory=libs/skinny

%sh curl -LsSf https://raw.githubusercontent.com/mlflow/mlflow/HEAD/dev/install-skinny.sh | sh -s pull/18671/merge

NameError                                 Traceback (most recent call last)
Cell In[59], line 2
      1 # Load the model back for predictions as a generic Python Function model
----> 2 loaded_model = mlflow.pyfunc.load_model(model_info.model_uri)
      4 predictions = loaded_model.predict(X_test)
      6 iris_feature_names = datasets.load_iris().feature_names

NameError: name 'model_info' is not defined

Downloading artifacts: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 12.20it/s]
---------------------------------------------------------------------------
MlflowException                           Traceback (most recent call last)
Cell In[58], line 2
      1 # Load the model back for predictions as a generic Python Function model
----> 2 loaded_model = mlflow.pyfunc.load_model("mlflow-artifacts:/916644062995257796/models/m-8cfa42826fdc479eb2512155cc91a54a/artifacts/MLmodel","iris_model")
      3 predictions = loaded_model.predict(X_test)
      5 iris_feature_names = datasets.load_iris().feature_names

File C:\Program Files\Python313\Lib\site-packages\mlflow\tracing\provider.py:567, in trace_disabled.<locals>.wrapper(*args, **kwargs)
    565 try:
    566     is_func_called = True
--> 567     result = f(*args, **kwargs)
    568 finally:
    569     enable()

File C:\Program Files\Python313\Lib\site-packages\mlflow\pyfunc\__init__.py:1140, in load_model(model_uri, suppress_warnings, dst_path, model_config)
   1137     model_requirements = _get_pip_requirements_from_model_path(local_path)
   1138     warn_dependency_requirement_mismatches(model_requirements)
-> 1140 model_meta = Model.load(os.path.join(local_path, MLMODEL_FILE_NAME))
   1142 if model_meta.metadata and model_meta.metadata.get(MLFLOW_MODEL_IS_EXTERNAL, False) is True:
   1143     raise MlflowException(
   1144         "This model's artifacts are external and are not stored in the model directory."
   1145         " This model cannot be loaded with MLflow.",
   1146         BAD_REQUEST,
   1147     )

File C:\Program Files\Python313\Lib\site-packages\mlflow\models\model.py:814, in Model.load(cls, path)
    812 is_model_dir = path.rsplit(sep, maxsplit=1)[-1] != MLMODEL_FILE_NAME
    813 mlmodel_file_path = f"{path}/{MLMODEL_FILE_NAME}" if is_model_dir else path
--> 814 mlmodel_local_path = _download_artifact_from_uri(artifact_uri=mlmodel_file_path)
    815 with open(mlmodel_local_path) as f:
    816     model_dict = yaml.safe_load(f)

File C:\Program Files\Python313\Lib\site-packages\mlflow\tracking\artifact_utils.py:117, in _download_artifact_from_uri(artifact_uri, output_path, lineage_header_info, tracking_uri, registry_uri)
    107 """
    108 Args:
    109     artifact_uri: The *absolute* URI of the artifact to download.
   (...)    114     registry_uri: The registry URI to be used when downloading artifacts.
    115 """
    116 root_uri, artifact_path = _get_root_uri_and_artifact_path(artifact_uri)
--> 117 repo = get_artifact_repository(
    118     artifact_uri=root_uri, tracking_uri=tracking_uri, registry_uri=registry_uri
    119 )
    121 try:
    122     if isinstance(repo, ModelsArtifactRepository):

File C:\Program Files\Python313\Lib\site-packages\mlflow\store\artifact\artifact_repository_registry.py:147, in get_artifact_repository(arti...

</details>

- Fixes mlflow/mlflow#18670

<!-- START COPILOT CODING AGENT TIPS -->
---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our [2 minute survey](https://gh.io/copilot-coding-agent-survey).