➖ Drop support for python 3.7 (#391)

.github/workflows/test.yml

@@ -14,7 +14,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        python-version: [3.7, 3.8]
+        python-version: ["3.8", "3.9", "3.10"]
         os: [ubuntu-latest, macos-latest, windows-latest]
     env:
       OS: ${{ matrix.os }}
@@ -30,15 +30,15 @@ jobs:
         python -m pip install --upgrade pip
         pip install .[test,extras]
     - name: Check code formatting with black
-      if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.7' # linting should occur only once in the loop
+      if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.10' # linting should occur only once in the loop
       run: |
         black . --check
     - name: Check import order with isort
-      if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.7' # linting should occur only once in the loop
+      if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.10' # linting should occur only once in the loop
       run: |
         isort . --check-only
     - name: Lint with flake8
-      if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.7' # linting should occur only once in the loop
+      if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.10' # linting should occur only once in the loop
       run: |
         # stop the build if there are Python syntax errors or undefined names
         flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics --exclude kedro_mlflow/template/project/run.py
@@ -49,7 +49,7 @@ jobs:
         pytest --cov=./ --cov-report=xml
     - name: Upload coverage report to Codecov
       uses: codecov/codecov-action@v1
-      if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.7' # upload should occur only once in the loop
+      if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.10' # upload should occur only once in the loop
       with:
         token: ${{ secrets.CODECOV_TOKEN }}  # token is not mandatory but make access more stable
         file: ./coverage.xml

CHANGELOG.md

@@ -5,6 +5,10 @@
 
 
 -   :bug: Make ``kedro-mlflow`` hook log parameters when the project is configured with the ``OmegaConfigLoader`` instead of raising an error ([#430](https://github.com/Galileo-Galilei/kedro-mlflow/issues/430))
+### Removed
+
+
+-   :bug: Drop support for ``python=3.7`` which has [reached end-of-life status](https://devguide.python.org/versions/) to prepare 0.19 ([#391](https://github.com/Galileo-Galilei/kedro-mlflow/issues/391))
 
 ## [0.11.8] - 2023-02-13
 

CONTRIBUTING.md

@@ -7,7 +7,7 @@ The current workflow is the following:
 3. Develop locally:
     - Install the precommit file (`pip install pre-commit`, then `pre-commit install`)
     - Create a branch based on the master branch (``git checkout -b <prefix-branchname> master``)
-    - Create a conda environment (conda create -n <your-env-name> python==3.7)
+    - Create a conda environment (conda create -n <your-env-name> python==3.10)
     - Activate this environment (`conda activate <your-env-name>`)
     - Install the extra dependencies for tests (`pip install kedro-mlflow[dev,test]`)
     - Apply your changes

docs/source/05_framework_ml/03_framework_solutions.md

@@ -70,7 +70,7 @@ mlflow.pyfunc.log_model(
     artifact_path="model",
     python_model=kedro_model,
     artifacts=artifacts,
-    conda_env={"python": "3.7.0", dependencies: ["kedro==0.16.5"]},
+    conda_env={"python": "3.10.0", dependencies: ["kedro==0.18.11"]},
     signature=model_signature,
 )
 ```

docs/source/05_pipeline_serving/02_custom_kedro_pipeline_model.md

@@ -37,7 +37,7 @@ mlflow.pyfunc.log_model(
     artifact_path="model",
     python_model=kedro_pipeline_model,
     artifacts=artifacts,
-    conda_env={"python": "3.7.0", dependencies: ["kedro==0.16.5"]},
+    conda_env={"python": "3.10.0", dependencies: ["kedro==0.18.11"]},
     model_signature=model_signature,
 )
 ```

docs/source/05_pipeline_serving/04_hook_pipeline_ml.md

@@ -23,7 +23,7 @@ For consistency, you may want to log an inference pipeline (including some data
             log_model_kwargs=dict(
                 artifact_path="kedro_mlflow_tutorial",
                 conda_env={
-                    "python": 3.7,
+                    "python": 3.10,
                     "dependencies": [f"kedro_mlflow_tutorial=={PROJECT_VERSION}"],
                 },
                 signature="auto",

docs/source/07_python_objects/01_DataSets.md

@@ -94,7 +94,7 @@ mlflow_model_logger = MlflowModelLoggerDataSet(
     flavor="mlflow.sklearn",
     run_id="<the-model-run-id>",
     save_args={
-        "conda_env": {"python": "3.7.0", "dependencies": ["kedro==0.16.5"]},
+        "conda_env": {"python": "3.10.0", "dependencies": ["kedro==0.18.11"]},
         "input_example": data.iloc[0:5, :],
     },
 )
@@ -110,9 +110,9 @@ my_model:
     run_id: <the-model-run-id>,
     save_args:
         conda_env:
-            python: "3.7.0"
+            python: "3.10.0"
             dependencies:
-                - "kedro==0.16.5"
+                - "kedro==0.18.11"
 ```
 
 ### ``MlflowModelSaverDataSet``

docs/source/07_python_objects/03_Pipelines.md

@@ -61,7 +61,7 @@ mlflow.pyfunc.log_model(
     artifact_path="model",
     python_model=KedroPipelineModel(pipeline=pipeline_training, catalog=catalog),
     artifacts=artifacts,
-    conda_env={"python": "3.7.0", "dependencies": ["kedro==0.16.5"]},
+    conda_env={"python": "3.10.0", dependencies: ["kedro==0.18.11"]},
     signature=model_signature,
 )
 ```

kedro_mlflow/framework/hooks/mlflow_hook.py

@@ -89,22 +89,15 @@ def after_context_created(
             mlflow_config.server._mlflow_client = MlflowClient(
                 tracking_uri=mlflow_config.server.mlflow_tracking_uri
             )
-            # BEWARE: kedro supports python=3.7 which does nto accepts the shorthand f"{x=}" for f"x={x}"
-            LOGGER.warning(
-                f"mlflow_config.server.mlflow_tracking_uri={mlflow_config.server.mlflow_tracking_uri}"
-            )
+            LOGGER.warning(f"{mlflow_config.server.mlflow_tracking_uri=}")
 
             mlflow_config.tracking.run.id = active_run_info.run_id
-            LOGGER.warning(
-                f"mlflow_config.tracking.run.id={mlflow_config.tracking.run.id}"
-            )
+            LOGGER.warning(f"{mlflow_config.tracking.run.id=}")
 
             mlflow_config.tracking.experiment.name = mlflow.get_experiment(
                 experiment_id=active_run_info.experiment_id
             ).name
-            LOGGER.warning(
-                f"mlflow_config.tracking.experiment.name={mlflow_config.tracking.experiment.name}"
-            )
+            LOGGER.warning(f"{mlflow_config.tracking.experiment.name=}")
 
         else:
             # we infer and setup the configuration only if there is no active run:

setup.py

@@ -7,7 +7,7 @@
 
 
 def _parse_requirements(path, encoding="utf-8"):
-    with open(path, mode="r", encoding=encoding) as file_handler:
+    with open(path, encoding=encoding) as file_handler:
         requirements = [
             x.strip() for x in file_handler if x.strip() and not x.startswith("-r")
         ]
@@ -31,7 +31,7 @@ def _parse_requirements(path, encoding="utf-8"):
     long_description=README,
     long_description_content_type="text/markdown",
     url="https://github.com/Galileo-Galilei/kedro-mlflow",
-    python_requires=">=3.7, <3.11",
+    python_requires=">=3.8, <3.11",
     packages=find_packages(exclude=["docs*", "tests*"]),
     setup_requires=["setuptools_scm"],
     include_package_data=True,
@@ -73,7 +73,6 @@ def _parse_requirements(path, encoding="utf-8"):
     keywords="kedro-plugin, mlflow, model versioning, model packaging, pipelines, machine learning, data pipelines, data science, data engineering",
     classifiers=[
         "Development Status :: 4 - Beta",
-        "Programming Language :: Python :: 3.7",
         "Programming Language :: Python :: 3.8",
         "Programming Language :: Python :: 3.9",
         "Programming Language :: Python :: 3.10",

tests/framework/hooks/test_hook_pipeline_ml.py

@@ -147,9 +147,9 @@ def convert_probs_to_pred(data, threshold):
         training=full_pipeline.only_nodes_with_tags("training"),
         inference=full_pipeline.only_nodes_with_tags("inference"),
         input_name="data",
-        log_model_kwargs={
-            "conda_env": {"python": "3.7.0", "dependencies": ["kedro==0.16.5"]},
-        },
+        log_model_kwargs=dict(
+            conda_env=dict(python="3.10.0", dependencies=["kedro==0.18.11"])
+        ),
     )
     return pipeline_ml_with_parameters
 
@@ -300,7 +300,7 @@ def test_mlflow_hook_save_pipeline_ml_with_copy_mode(
             input_name=dummy_pipeline_ml.input_name,
             log_model_kwargs={
                 "artifact_path": dummy_pipeline_ml.log_model_kwargs["artifact_path"],
-                "conda_env": {"python": "3.7.0", "dependencies": ["kedro==0.16.5"]},
+                "conda_env": {"python": "3.10.0", "dependencies": ["kedro==0.18.11"]},
             },
             kpm_kwargs={
                 "copy_mode": copy_mode,

tests/io/models/test_mlflow_model_logger_dataset.py

@@ -114,7 +114,6 @@ def dummy_catalog(tmp_path):
 
 @pytest.fixture
 def kedro_pipeline_model(pipeline_ml_obj, dummy_catalog):
-
     kedro_pipeline_model = KedroPipelineModel(
         pipeline=pipeline_ml_obj,
         catalog=dummy_catalog,
@@ -175,7 +174,6 @@ def test_save_sklearn_flavor_with_run_id_and_already_active_run(tracking_uri):
 def test_save_and_load_sklearn_flavor_with_run_id(
     tracking_uri, mlflow_client, linreg_model, active_run_when_loading
 ):
-
     mlflow.set_tracking_uri(tracking_uri)
     # close all opened mlflow runs to avoid interference between tests
     while mlflow.active_run():
@@ -217,7 +215,6 @@ def test_save_and_load_sklearn_flavor_with_run_id(
 def test_save_and_load_sklearn_flavor_without_run_id(
     tracking_uri, mlflow_client, linreg_model, initial_active_run
 ):
-
     mlflow.set_tracking_uri(tracking_uri)
     # close all opened mlflow runs to avoid interference between tests
     while mlflow.active_run():
@@ -265,7 +262,6 @@ def test_save_and_load_sklearn_flavor_without_run_id(
 
 
 def test_load_without_run_id_nor_active_run(tracking_uri):
-
     mlflow.set_tracking_uri(tracking_uri)
     # close all opened mlflow runs to avoid interference between tests
     while mlflow.active_run():
@@ -303,7 +299,6 @@ def test_pyfunc_flavor_python_model_save_and_load(
     pipeline,
     dummy_catalog,
 ):
-
     kedro_pipeline_model = KedroPipelineModel(
         pipeline=pipeline,
         catalog=dummy_catalog,
@@ -320,7 +315,7 @@ def test_pyfunc_flavor_python_model_save_and_load(
             "artifact_path": "test_model",
             "save_args": {
                 "artifacts": artifacts,
-                "conda_env": {"python": "3.7.0", "dependencies": ["kedro==0.16.5"]},
+                "conda_env": {"python": "3.10.0", "dependencies": ["kedro==0.18.11"]},
             },
         },
     }
@@ -348,7 +343,6 @@ def test_pyfunc_flavor_python_model_save_and_load(
 
 
 def test_pyfunc_flavor_wrong_pyfunc_workflow(tracking_uri):
-
     model_config = {
         "name": "kedro_pipeline_model",
         "config": {
@@ -373,23 +367,19 @@ def test_mlflow_model_logger_logging_deactivation(tracking_uri, linreg_model):
 
     mlflow_model_logger_dataset._logging_activated = False
 
-    all_runs_id_beginning = set(
-        [
-            run.run_id
-            for k in range(len(mlflow_client.search_experiments()))
-            for run in mlflow_client.search_runs(experiment_ids=f"{k}")
-        ]
-    )
+    all_runs_id_beginning = {
+        run.run_id
+        for k in range(len(mlflow_client.search_experiments()))
+        for run in mlflow_client.search_runs(experiment_ids=f"{k}")
+    }
 
     mlflow_model_logger_dataset.save(linreg_model)
 
-    all_runs_id_end = set(
-        [
-            run.run_id
-            for k in range(len(mlflow_client.search_experiments()))
-            for run in mlflow_client.search_runs(experiment_ids=f"{k}")
-        ]
-    )
+    all_runs_id_end = {
+        run.run_id
+        for k in range(len(mlflow_client.search_experiments()))
+        for run in mlflow_client.search_runs(experiment_ids=f"{k}")
+    }
 
     assert all_runs_id_beginning == all_runs_id_end
 

tests/io/models/test_mlflow_model_saver_dataset.py

@@ -94,7 +94,6 @@ def dummy_catalog(tmp_path):
 
 @pytest.fixture
 def kedro_pipeline_model(tmp_path, pipeline_ml_obj, dummy_catalog):
-
     kedro_pipeline_model = KedroPipelineModel(
         pipeline=pipeline_ml_obj,
         catalog=dummy_catalog,
@@ -158,7 +157,6 @@ def test_save_load_local(linreg_path, linreg_model, versioned):
 def test_pyfunc_flavor_python_model_save_and_load(
     tmp_path, tmp_folder, pipeline, dummy_catalog
 ):
-
     kedro_pipeline_model = KedroPipelineModel(
         pipeline=pipeline,
         catalog=dummy_catalog,
@@ -177,7 +175,7 @@ def test_pyfunc_flavor_python_model_save_and_load(
             "pyfunc_workflow": "python_model",
             "save_args": {
                 "artifacts": artifacts,
-                "conda_env": {"python": "3.7.0", "dependencies": ["kedro==0.16.5"]},
+                "conda_env": {"python": "3.10.0", "dependencies": ["kedro==0.18.11"]},
             },
         },
     }

tests/mlflow/test_kedro_pipeline_model.py

@@ -263,7 +263,6 @@ def dummy_catalog(tmp_path):
 def test_model_packaging_with_copy_mode(
     tmp_path, tmp_folder, pipeline_inference_dummy, dummy_catalog, copy_mode, expected
 ):
-
     dummy_catalog._data_sets["model"].save(2)  # emulate model fitting
 
     kedro_model = KedroPipelineModel(
@@ -282,7 +281,7 @@ def test_model_packaging_with_copy_mode(
             artifact_path="model",
             python_model=kedro_model,
             artifacts=artifacts,
-            conda_env={"python": "3.7.0", "dependencies": ["kedro==0.16.5"]},
+            conda_env={"python": "3.10.0", "dependencies": ["kedro==0.18.11"]},
         )
         run_id = mlflow.active_run().info.run_id
 
@@ -320,7 +319,6 @@ def test_kedro_pipeline_model_with_wrong_copy_mode_type(
 
 
 def test_model_packaging_too_many_artifacts(tmp_path, pipeline_inference_dummy):
-
     catalog = DataCatalog(
         {
             "raw_data": PickleDataSet(
@@ -356,7 +354,7 @@ def test_model_packaging_too_many_artifacts(tmp_path, pipeline_inference_dummy):
             artifact_path="model",
             python_model=kedro_model,
             artifacts=artifacts,
-            conda_env={"python": "3.7.0", "dependencies": ["kedro==0.16.5"]},
+            conda_env={"python": "3.10.0", "dependencies": ["kedro==0.18.11"]},
         )
         run_id = mlflow.active_run().info.run_id
 
@@ -369,7 +367,6 @@ def test_model_packaging_too_many_artifacts(tmp_path, pipeline_inference_dummy):
 
 
 def test_model_packaging_missing_artifacts(tmp_path, pipeline_inference_dummy):
-
     catalog = DataCatalog(
         {
             "raw_data": MemoryDataSet(),
@@ -391,7 +388,7 @@ def test_model_packaging_missing_artifacts(tmp_path, pipeline_inference_dummy):
             artifact_path="model",
             python_model=kedro_model,
             artifacts=None,  # no artifacts provided
-            conda_env={"python": "3.7.0", "dependencies": ["kedro==0.16.5"]},
+            conda_env={"python": "3.10.0", "dependencies": ["kedro==0.18.11"]},
         )
         run_id = mlflow.active_run().info.run_id
 
@@ -404,7 +401,6 @@ def test_model_packaging_missing_artifacts(tmp_path, pipeline_inference_dummy):
 
 
 def test_kedro_pipeline_ml_loading_deepcopiable_catalog(tmp_path, tmp_folder):
-
     # create pipelien and catalog. The training will not be triggered
     def fit_fun(data):
         pass
@@ -510,7 +506,6 @@ def test_catalog_extraction(pipeline, catalog, input_name, result):
 
 
 def test_catalog_extraction_missing_inference_input(pipeline_inference_dummy):
-
     catalog = DataCatalog({"raw_data": MemoryDataSet(), "data": MemoryDataSet()})
     # "model" is missing in the catalog
     with pytest.raises(
@@ -599,7 +594,6 @@ def test_kedro_pipeline_model_save_and_load(
 
 
 def test_kedro_pipeline_model_too_many_outputs():
-
     catalog = DataCatalog(
         {
             "data": MemoryDataSet(),