FIX #116 - Pass filepath in MlflowAbstractModelDataSet through PurePo…

…sixPath in init

.pre-commit-config.yaml

@@ -1,7 +1,7 @@
 exclude: ^kedro_mlflow/template/project/run.py$
 repos:
     - repo: https://github.com/psf/black
-      rev: 19.10b0
+      rev: 20.8b1
       hooks:
           - id: black
             language_version: python3.7
@@ -10,11 +10,11 @@ repos:
       hooks:
           - id: isort
     - repo: https://gitlab.com/pycqa/flake8
-      rev: 3.7.9
+      rev: 3.8.4
       hooks:
           - id: flake8
     - repo: https://github.com/pre-commit/pre-commit-hooks
-      rev: v2.5.0
+      rev: v3.3.0
       hooks:
           - id: check-merge-conflict
           - id: debug-statements

CHANGELOG.md

@@ -2,6 +2,10 @@
 
 ## [Unreleased]
 
+### Fixed
+
+-   Fix `TypeError: unsupported operand type(s) for /: 'str' and 'str'` when using `MlflowArtifactDataSet` with `MlflowModelSaverDataSet` ([#116](https://github.com/Galileo-Galilei/kedro-mlflow/issues/116))
+
 ## [0.4.0] - 2020-11-03
 
 ### Added

docs/source/03_tutorial/06_version_models.md

@@ -51,3 +51,18 @@ my_custom_model:
     flavor: my_package.custom_mlflow_flavor
     pyfunc_workflow: python_model # or loader_module
 ```
+
+### Hwo can I save model locally and log it in MLflow in one step?
+
+If you want to save your model both locally and remotely within the same run, you can leverage `MlflowArtifactDataSet`:
+
+```yaml
+sklearn_model:
+    type: kedro_mlflow.io.artifacts.MlflowArtifactDataSet
+    data_set:
+        type: kedro_mlflow.io.models.MlflowModelSaverDataSet
+        flavor: mlflow.sklearn
+        filepath: data/06_models/sklearn_model
+```
+
+This might be useful if you want to always read the lastest model saved locally and log it to MLflow each time the new model is being trained for tracking purpose.

kedro_mlflow/io/models/mlflow_abstract_model_dataset.py

@@ -1,8 +1,9 @@
 import importlib
+from pathlib import PurePosixPath
 from typing import Any, Dict, Optional
 
 from kedro.io import AbstractVersionedDataSet, Version
-from kedro.io.core import DataSetError
+from kedro.io.core import DataSetError, get_protocol_and_path
 
 
 class MlflowAbstractModelDataSet(AbstractVersionedDataSet):
@@ -12,7 +13,7 @@ class MlflowAbstractModelDataSet(AbstractVersionedDataSet):
 
     def __init__(
         self,
-        filepath,
+        filepath: str,
         flavor: str,
         pyfunc_workflow: Optional[str] = None,
         load_args: Dict[str, Any] = None,
@@ -27,23 +28,22 @@ def __init__(
         During load, the model is pulled from MLflow run with `run_id`.
 
         Args:
+            filepath (str): Path to store the dataset locally.
             flavor (str): Built-in or custom MLflow model flavor module.
                 Must be Python-importable.
-            filepath (str): Path to store the dataset locally.
-            run_id (Optional[str], optional): MLflow run ID to use to load
-                the model from or save the model to. If provided,
-                takes precedence over filepath. Defaults to None.
             pyfunc_workflow (str, optional): Either `python_model` or `loader_module`.
                 See https://www.mlflow.org/docs/latest/python_api/mlflow.pyfunc.html#workflows.
             load_args (Dict[str, Any], optional): Arguments to `load_model`
                 function from specified `flavor`. Defaults to {}.
             save_args (Dict[str, Any], optional): Arguments to `log_model`
                 function from specified `flavor`. Defaults to {}.
+            version (Version, optional): Specific version to load.
 
         Raises:
             DataSetError: When passed `flavor` does not exist.
         """
-        super().__init__(filepath, version)
+        _, path = get_protocol_and_path(filepath, version)
+        super().__init__(PurePosixPath(path), version)
         self._flavor = flavor
         self._pyfunc_workflow = pyfunc_workflow
 

kedro_mlflow/io/models/mlflow_model_saver_dataset.py

@@ -27,7 +27,7 @@ def __init__(
 
         Parameters are passed from the Data Catalog.
 
-        During save, the model is saved locally at `filepat`
+        During save, the model is saved locally at `filepath`
         During load, the model is loaded from the local `filepath`.
 
         Args:

tests/io/models/test_mlflow_model_saver_dataset.py

@@ -91,14 +91,15 @@ def kedro_pipeline_model(tmp_path, pipeline_ml_obj, dummy_catalog):
 
 
 def test_save_unversioned_under_same_path(
-    linreg_path, linreg_model,
+    linreg_path,
+    linreg_model,
 ):
     model_config = {
         "name": "linreg",
         "config": {
             "type": "kedro_mlflow.io.models.MlflowModelSaverDataSet",
             "flavor": "mlflow.sklearn",
-            "filepath": linreg_path,
+            "filepath": linreg_path.as_posix(),
         },
     }
     mlflow_model_ds = MlflowModelSaverDataSet.from_config(**model_config)
@@ -114,7 +115,7 @@ def test_save_load_local(linreg_path, linreg_model, versioned):
         "name": "linreg",
         "config": {
             "type": "kedro_mlflow.io.models.MlflowModelSaverDataSet",
-            "filepath": linreg_path,
+            "filepath": linreg_path.as_posix(),
             "flavor": "mlflow.sklearn",
             "versioned": versioned,
         },
@@ -143,7 +144,9 @@ def test_pyfunc_flavor_python_model_save_and_load(
         "name": "kedro_pipeline_model",
         "config": {
             "type": "kedro_mlflow.io.models.MlflowModelSaverDataSet",
-            "filepath": tmp_path / "data" / "06_models" / "my_custom_model",
+            "filepath": (
+                tmp_path / "data" / "06_models" / "my_custom_model"
+            ).as_posix(),
             "flavor": "mlflow.pyfunc",
             "pyfunc_workflow": "python_model",
             "save_args": {"artifacts": artifacts, "conda_env": {"python": "3.7.0"}},