You can load python_function models durante Python by calling the mlflow

pyfunc.load_model() function. Note that the load_model function assumes that all dependencies are already available and will not check nor install any dependencies ( see model deployment section for tools sicuro deploy models with automatic dependency management).

All PyFunc models will support pandas.DataFrame as an spinta. Sopra prime sicuro pandas.DataFrame , DL PyFunc models will also support tensor inputs sopra the form of numpy.ndarrays . To verify whether a model flavor supports tensor inputs, please check the flavor’s documentation.

For models with per column-based lista, inputs are typically provided durante the form of verso pandas.DataFrame . If verso dictionary mapping column name to values is provided as spinta for schemas with named columns or if per python List or a numpy.ndarray is provided as incentivo for schemas with unnamed columns, MLflow will cast the input preciso per DataFrame. Specifica enforcement and casting with respect esatto the expected tempo types is performed against the DataFrame.

For models with verso tensor-based schema, inputs are typically provided durante the form of per numpy.ndarray or per dictionary mapping the tensor name sicuro its np.ndarray value. Schema enforcement will check the provided input’s shape and type against the shape and type specified con the model’s specifica and throw an error if they do not gara.

For models where niente affatto specifica is defined, giammai changes puro the model inputs and outputs are made. MLflow will propogate any errors raised by the model if the model does not accept the provided incentivo type.

R Function ( crate )

The crate model flavor defines verso generic model format for representing an arbitrary R prediction function as an MLflow model using the crate function from the carrier package. The prediction function is expected preciso take per dataframe as stimolo and produce per dataframe, verso vector or a list with the predictions as output.

H2O ( h2o )

The mlflow.h2o ondoie defines save_model() and log_model() methods per python, and mlflow_save_model and mlflow_log_model sopra R for saving H2O models mediante MLflow Model format. These methods produce MLflow Models with the python_function flavor, allowing you preciso load them as generic Python functions for inference modo mlflow.pyfunc.load_model() . This loaded PyFunc model can be scored with only DataFrame incentivo. When you load MLflow Models with the h2o flavor using mlflow.pyfunc.load_model() , the h2o.init() method is called. Therefore, the correct version of h2o(-py) must be installed sopra the loader’s environment. You can customize the arguments given puro h2o.init() by modifying the init entry of the persisted H2O model’s YAML configuration file: model.h2o/h2o.yaml .

Keras ( keras )

The keras model flavor enables logging and loading Keras models. It is available mediante both Python and R clients. The mlflow.keras diversifie defines save_model() and log_model() functions that you can use to save Keras models per MLflow Model format in Python. Similarly, con R, you can save or log the model using mlflow_save_model and mlflow_log_model. These functions serialize Keras models as HDF5 files using the Keras library’s built-sopra model persistence functions. MLflow Models produced by these functions also contain the python_function flavor, allowing them onesto be interpreted as generic Python functions for inference modo mlflow.pyfunc.load_model() . This loaded PyFunc model can be scored with both DataFrame spinta and numpy array stimolo. Finally, you can use the mlflow.keras.load_model() function mediante Python or mlflow_load_model function per R puro load MLflow Models with the keras flavor as Keras Model objects.

MLeap ( mleap )

The mleap model flavor supports saving Spark models mediante MLflow format using the MLeap persistence mechanism. MLeap is an inference-optimized format and execution engine for Spark models that does not depend on SparkContext to evaluate inputs.