Merge branch 'main' of github.com:fmind/mlops-python-package

Author: fmind

.github/ISSUE_TEMPLATE/feat-request.md

@@ -4,15 +4,6 @@ about: A new feature.
 title: "[FEAT] "
 labels: feat
 assignees: fmind
-body:
-- type: checkboxes
-  id: breaking-change
-  attributes:
-    label: Breaking change?
-    options:
-      - label: Yes
-      - label: No
-      - label: Maybe
 ---
 
 ## Description

.github/ISSUE_TEMPLATE/fix-request.md

@@ -4,15 +4,6 @@ about: A bug fix
 title: "[FIX] "
 labels: fix
 assignees: fmind
-body:
-- type: checkboxes
-  id: breaking-change
-  attributes:
-    label: Breaking change?
-    options:
-      - label: Yes
-      - label: No
-      - label: Maybe
 ---
 
 ## Bug Description

.github/workflows/publish.yml

@@ -3,7 +3,7 @@ on:
   release:
     types: [published]
 env:
-  DOCKER_IMAGE: ghcr.io/fmind/mlops-python-package:latest
+  DOCKER_IMAGE: ghcr.io/fmind/mlops-python-package
 concurrency:
   cancel-in-progress: true
   group: publish-workflow
@@ -39,4 +39,6 @@ jobs:
           context: .
           cache-to: type=gha
           cache-from: type=gha
-          tags: ${{ env.DOCKER_IMAGE }}
+          tags: |
+            ${{ env.DOCKER_IMAGE }}:latest
+            ${{ env.DOCKER_IMAGE }}:${{ github.ref }}

pyproject.toml

@@ -92,6 +92,7 @@ addopts = "--verbosity=2"
 
 [tool.ruff]
 fix = true
+indent-width = 4
 line-length = 100
 target-version = "py312"
 

src/bikes/core/schemas.py

@@ -11,6 +11,7 @@
 
 # %% TYPES
 
+# Generic type for a dataframe container
 TSchema = T.TypeVar("TSchema", bound="pa.DataFrameModel")
 
 # %% SCHEMAS

src/bikes/io/services.py

@@ -121,7 +121,7 @@ class RunConfig(pdt.BaseModel, strict=True, frozen=True, extra="forbid"):
     experiment_name: str = "bikes"
     # registry
     registry_name: str = "bikes"
-    # autologg
+    # autolog
     autolog_disable: bool = False
     autolog_disable_for_unsupported_versions: bool = False
     autolog_exclusive: bool = False
@@ -138,7 +138,7 @@ def start(self) -> None:
         mlflow.set_registry_uri(uri=self.registry_uri)
         # experiment
         mlflow.set_experiment(experiment_name=self.experiment_name)
-        # autologging
+        # autolog
         mlflow.autolog(
             disable=self.autolog_disable,
             disable_for_unsupported_versions=self.autolog_disable_for_unsupported_versions,

tests/conftest.py

@@ -78,8 +78,8 @@ def extra_config() -> str:
     {
         "job": {
             "mlflow_service": {
-                "tracking_uri": "${tmp_path:}/experiments/",
-                "registry_uri": "${tmp_path:}/models/",
+                "tracking_uri": "${tmp_path:}/tracking/",
+                "registry_uri": "${tmp_path:}/registry/",
             }
         }
     }

tests/io/test_registries.py

@@ -43,7 +43,7 @@ def test_custom_pipeline(
     saver = registries.CustomSaver(path=path)
     loader = registries.CustomLoader()
     register = registries.MlflowRegister(tags=tags)
-    run_config = services.MlflowService.RunConfig(name="Custom-Run")
+    run_config = mlflow_service.RunConfig(name="Custom-Run")
     # when
     with mlflow_service.run_context(run_config=run_config) as run:
         info = saver.save(model=model, signature=signature, input_example=inputs)
@@ -92,7 +92,7 @@ def test_builtin_pipeline(
     saver = registries.BuiltinSaver(path=path, flavor=flavor)
     loader = registries.BuiltinLoader()
     register = registries.MlflowRegister(tags=tags)
-    run_config = services.MlflowService.RunConfig(name="Custom-Run")
+    run_config = mlflow_service.RunConfig(name="Builtin-Run")
     # when
     with mlflow_service.run_context(run_config=run_config) as run:
         info = saver.save(model=model, signature=signature, input_example=inputs)

tests/jobs/test_training.py

@@ -15,13 +15,13 @@ def test_training_job(
     targets_reader: datasets.Reader,
     model: models.Model,
     metric: metrics.Metric,
-    train_test_splitter: splitters.TrainTestSplitter,
+    train_test_splitter: splitters.Splitter,
     saver: registries.Saver,
     signer: signers.Signer,
     register: registries.Register,
 ) -> None:
     # given
-    run_config = services.MlflowService.RunConfig(
+    run_config = mlflow_service.RunConfig(
         name="TrainingTest", tags={"context": "training"}, description="Training job."
     )
     splitter = train_test_splitter

tests/jobs/test_tuning.py

@@ -15,11 +15,11 @@ def test_tuning_job(
     targets_reader: datasets.Reader,
     model: models.Model,
     metric: metrics.Metric,
-    time_series_splitter: splitters.TimeSeriesSplitter,
+    time_series_splitter: splitters.Splitter,
     searcher: searchers.Searcher,
 ) -> None:
     # given
-    run_config = services.MlflowService.RunConfig(
+    run_config = mlflow_service.RunConfig(
         name="TuningTest", tags={"context": "tuning"}, description="Tuning job."
     )
     splitter = time_series_splitter

tests/utils/test_searchers.py

@@ -11,7 +11,7 @@ def test_grid_cv_searcher(
     metric: metrics.Metric,
     inputs: schemas.Inputs,
     targets: schemas.Targets,
-    train_test_splitter: splitters.TrainTestSplitter,
+    train_test_splitter: splitters.Splitter,
 ) -> None:
     # given
     param_grid = {"max_depth": [3, 5, 7]}