hydraflow 0.6.2__tar.gz → 0.7.2__tar.gz

{hydraflow-0.6.2 → hydraflow-0.7.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: hydraflow
-Version: 0.6.2
+Version: 0.7.2
 Summary: Hydraflow integrates Hydra and MLflow to manage and track machine learning experiments.
 Project-URL: Documentation, https://daizutabi.github.io/hydraflow/
 Project-URL: Source, https://github.com/daizutabi/hydraflow

{hydraflow-0.6.2 → hydraflow-0.7.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "hydraflow"
-version = "0.6.2"
+version = "0.7.2"
 description = "Hydraflow integrates Hydra and MLflow to manage and track machine learning experiments."
 readme = "README.md"
 license = { file = "LICENSE" }
@@ -97,3 +97,4 @@ ignore = [
   "SLF",
 ]
 "apps/*.py" = ["D", "G", "INP"]
+"src/hydraflow/main.py" = ["ANN201", "D401", "PLR0913"]

{hydraflow-0.6.2 → hydraflow-0.7.2}/src/hydraflow/__init__.py

@@ -1,10 +1,11 @@
 """Integrate Hydra and MLflow to manage and track machine learning experiments."""
 
-from .config import select_config, select_overrides
-from .context import chdir_artifact, log_run, start_run
-from .mlflow import list_runs, search_runs, set_experiment
-from .run_collection import RunCollection
-from .utils import (
+from hydraflow.config import select_config, select_overrides
+from hydraflow.context import chdir_artifact, log_run, start_run
+from hydraflow.main import main
+from hydraflow.mlflow import list_runs, search_runs, set_experiment
+from hydraflow.run_collection import RunCollection
+from hydraflow.utils import (
     get_artifact_dir,
     get_artifact_path,
     get_hydra_output_dir,
@@ -25,6 +26,7 @@ __all__ = [
     "load_config",
     "load_overrides",
     "log_run",
+    "main",
     "remove_run",
     "search_runs",
     "select_config",

hydraflow-0.7.2/src/hydraflow/main.py

@@ -0,0 +1,55 @@
+"""main decorator."""
+
+from __future__ import annotations
+
+from functools import wraps
+from typing import TYPE_CHECKING, Any
+
+import hydra
+from hydra.core.config_store import ConfigStore
+from mlflow.entities import RunStatus
+
+import hydraflow
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from mlflow.entities import Run
+
+FINISHED = RunStatus.to_string(RunStatus.FINISHED)
+
+
+def main(
+    node: Any,
+    config_name: str = "config",
+    *,
+    chdir: bool = True,
+    skip_finished: bool = True,
+    force_new_run: bool = False,
+    override: bool = True,
+):
+    """Main decorator."""
+
+    def decorator(app: Callable[[Run, Any], None]) -> Callable[[], None]:
+        ConfigStore.instance().store(name=config_name, node=node)
+
+        @wraps(app)
+        @hydra.main(version_base=None, config_name=config_name)
+        def inner_app(cfg: object) -> None:
+            hydraflow.set_experiment()
+
+            if force_new_run:
+                run = None
+            else:
+                rc = hydraflow.search_runs()
+                run = rc.try_get(cfg, override=override)
+
+                if skip_finished and run and run.info.status == FINISHED:
+                    return
+
+            with hydraflow.start_run(cfg, run=run, chdir=chdir) as run:
+                app(run, cfg)
+
+        return inner_app
+
+    return decorator

{hydraflow-0.6.2 → hydraflow-0.7.2}/src/hydraflow/run_collection.py

@@ -286,105 +286,11 @@ class RunCollection:
             ),
         )
 
-    def find(self, config: object | None = None, **kwargs) -> Run:
-        """Find the first `Run` instance based on the provided configuration.
-
-        This method filters the runs in the collection according to the
-        specified configuration object and returns the first run that matches
-        the provided parameters. If no run matches the criteria, a `ValueError`
-        is raised.
-
-        Args:
-            config (object | None): The configuration object to identify the run.
-            **kwargs: Additional key-value pairs to filter the runs.
-
-        Returns:
-            The first `Run` instance that matches the provided configuration.
-
-        Raises:
-            ValueError: If no run matches the criteria.
-
-        See Also:
-            `filter`: Perform the actual filtering logic.
-
-        """
-        try:
-            return self.filter(config, **kwargs).first()
-        except ValueError:
-            raise ValueError("No run matches the provided configuration.")
-
-    def try_find(self, config: object | None = None, **kwargs) -> Run | None:
-        """Try to find the first `Run` instance based on the provided configuration.
-
-        This method filters the runs in the collection according to the
-        specified configuration object and returns the first run that matches
-        the provided parameters. If no run matches the criteria, None is
-        returned.
-
-        Args:
-            config (object | None): The configuration object to identify the run.
-            **kwargs: Additional key-value pairs to filter the runs.
-
-        Returns:
-            The first `Run` instance that matches the provided configuration, or
-            None if no runs match the criteria.
-
-        See Also:
-            `filter`: Perform the actual filtering logic.
-
-        """
-        return self.filter(config, **kwargs).try_first()
-
-    def find_last(self, config: object | None = None, **kwargs) -> Run:
-        """Find the last `Run` instance based on the provided configuration.
-
-        This method filters the runs in the collection according to the
-        specified configuration object and returns the last run that matches
-        the provided parameters. If no run matches the criteria, a `ValueError`
-        is raised.
-
-        Args:
-            config (object | None): The configuration object to identify the run.
-            **kwargs: Additional key-value pairs to filter the runs.
-
-        Returns:
-            The last `Run` instance that matches the provided configuration.
-
-        Raises:
-            ValueError: If no run matches the criteria.
-
-        See Also:
-            `filter`: Perform the actual filtering logic.
-
-        """
-        try:
-            return self.filter(config, **kwargs).last()
-        except ValueError:
-            raise ValueError("No run matches the provided configuration.")
-
-    def try_find_last(self, config: object | None = None, **kwargs) -> Run | None:
-        """Try to find the last `Run` instance based on the provided configuration.
-
-        This method filters the runs in the collection according to the
-        specified configuration object and returns the last run that matches
-        the provided parameters. If no run matches the criteria, None is
-        returned.
-
-        Args:
-            config (object | None): The configuration object to identify the run.
-            **kwargs: Additional key-value pairs to filter the runs.
-
-        Returns:
-            The last `Run` instance that matches the provided configuration, or
-            None if no runs match the criteria.
-
-        See Also:
-            `filter`: Perform the actual filtering logic.
-
-        """
-        return self.filter(config, **kwargs).try_last()
-
-    def get(self, config: object | None = None, **kwargs) -> Run:
+    def get(
+        self,
+        config: object | Callable[[Run], bool] | None = None,
+        **kwargs,
+    ) -> Run:
         """Retrieve a specific `Run` instance based on the provided configuration.
 
         This method filters the runs in the collection according to the
@@ -393,7 +299,10 @@ class RunCollection:
         one run matches the criteria, a `ValueError` is raised.
 
         Args:
-            config (object | None): The configuration object to identify the run.
+            config (object | Callable[[Run], bool] | None): The configuration object
+                to identify the run. This can be any object that provides key-value
+                pairs through the `iter_params` function, or a callable that
+                takes a `Run` object and returns a boolean value.
             **kwargs: Additional key-value pairs to filter the runs.
 
         Returns:
@@ -413,7 +322,11 @@ class RunCollection:
             msg = "The filtered collection does not contain exactly one run."
             raise ValueError(msg)
 
-    def try_get(self, config: object | None = None, **kwargs) -> Run | None:
+    def try_get(
+        self,
+        config: object | Callable[[Run], bool] | None = None,
+        **kwargs,
+    ) -> Run | None:
         """Try to get a specific `Run` instance based on the provided configuration.
 
         This method filters the runs in the collection according to the
@@ -422,7 +335,10 @@ class RunCollection:
         If more than one run matches the criteria, a `ValueError` is raised.
 
         Args:
-            config (object | None): The configuration object to identify the run.
+            config (object | Callable[[Run], bool] | None): The configuration object
+                to identify the run. This can be any object that provides key-value
+                pairs through the `iter_params` function, or a callable that
+                takes a `Run` object and returns a boolean value.
             **kwargs: Additional key-value pairs to filter the runs.
 
         Returns:

{hydraflow-0.6.2 → hydraflow-0.7.2}/tests/context/rerun.py

@@ -24,7 +24,7 @@ ConfigStore.instance().store(name="config", node=Config)
 def app(cfg: Config):
     hydraflow.set_experiment()
 
-    run = hydraflow.list_runs().try_find(cfg, override=True)
+    run = hydraflow.list_runs().try_get(cfg, override=True)
 
     with hydraflow.start_run(cfg, run=run) as run:
         log(hydraflow.get_artifact_dir(run))

hydraflow-0.7.2/tests/main/force_new_run.py

@@ -0,0 +1,26 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from pathlib import Path
+from typing import TYPE_CHECKING
+
+import hydraflow
+
+if TYPE_CHECKING:
+    from mlflow.entities import Run
+
+
+@dataclass
+class Config:
+    count: int = 0
+
+
+@hydraflow.main(Config, force_new_run=True)
+def app(run: Run, cfg: Config):
+    file = Path("a.txt")
+    text = file.read_text() if file.exists() else ""
+    file.write_text(text + f"{cfg.count}")
+
+
+if __name__ == "__main__":
+    app()

hydraflow-0.7.2/tests/main/restart.py

@@ -0,0 +1,26 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from pathlib import Path
+from typing import TYPE_CHECKING
+
+import hydraflow
+
+if TYPE_CHECKING:
+    from mlflow.entities import Run
+
+
+@dataclass
+class Config:
+    count: int = 0
+
+
+@hydraflow.main(Config, skip_finished=False)
+def app(run: Run, cfg: Config):
+    file = Path("a.txt")
+    text = file.read_text() if file.exists() else ""
+    file.write_text(text + f"{cfg.count}")
+
+
+if __name__ == "__main__":
+    app()

hydraflow-0.7.2/tests/main/skip.py

@@ -0,0 +1,26 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from pathlib import Path
+from typing import TYPE_CHECKING
+
+import hydraflow
+
+if TYPE_CHECKING:
+    from mlflow.entities import Run
+
+
+@dataclass
+class Config:
+    count: int = 0
+
+
+@hydraflow.main(Config)
+def app(run: Run, cfg: Config):
+    file = Path("a.txt")
+    text = file.read_text() if file.exists() else ""
+    file.write_text(text + f"{cfg.count} {run.info.run_id}\n")
+
+
+if __name__ == "__main__":
+    app()

hydraflow-0.7.2/tests/main/test_force_new_run.py

@@ -0,0 +1,33 @@
+import pytest
+from mlflow.entities import Run
+
+from hydraflow.run_collection import RunCollection
+
+pytestmark = pytest.mark.xdist_group(name="group6")
+
+
+@pytest.fixture(scope="module")
+def rc(collect):
+    for _ in range(3):
+        rc = collect("main/force_new_run.py", ["count=3"])
+    return rc
+
+
+def test_rc_len(rc: RunCollection):
+    assert len(rc) == 3
+
+
+def test_rc_filter(rc: RunCollection):
+    assert len(rc.filter(count=3)) == 3
+
+
+@pytest.fixture(scope="module", params=range(3))
+def run(rc: RunCollection, request: pytest.FixtureRequest):
+    return rc[request.param]
+
+
+def test_count(run: Run):
+    from hydraflow.utils import get_artifact_path
+
+    path = get_artifact_path(run, "a.txt")
+    assert path.read_text() == "3"

hydraflow-0.7.2/tests/main/test_restart.py

@@ -0,0 +1,24 @@
+import pytest
+
+from hydraflow.run_collection import RunCollection
+
+pytestmark = pytest.mark.xdist_group(name="group7")
+
+
+@pytest.fixture(scope="module")
+def rc(collect):
+    for _ in range(3):
+        rc = collect("main/restart.py", ["count=3"])
+    return rc
+
+
+def test_rc_len(rc: RunCollection):
+    assert len(rc) == 1
+
+
+def test_count(rc: RunCollection):
+    from hydraflow.utils import get_artifact_path
+
+    run = rc.get(count=3)
+    path = get_artifact_path(run, "a.txt")
+    assert path.read_text() == "333"

hydraflow-0.6.2/tests/context/test_preemption.py → hydraflow-0.7.2/tests/main/test_skip.py

@@ -12,7 +12,7 @@ def rc(collect):
     client = MlflowClient()
     running = RunStatus.to_string(RunStatus.RUNNING)
 
-    filename = "context/preemption.py"
+    filename = "main/skip.py"
     args = ["-m", "count=1,2,3"]
 
     rc = collect(filename, args)
@@ -32,10 +32,27 @@ def run(rc: RunCollection, request: pytest.FixtureRequest):
     return rc.get(count=request.param)
 
 
-def test_run_count(run: Run):
+@pytest.fixture(scope="module")
+def count(run: Run):
+    return int(run.data.params["count"])
+
+
+@pytest.fixture(scope="module")
+def text(run: Run):
     from hydraflow.utils import get_artifact_path
 
-    count = int(run.data.params["count"])
     path = get_artifact_path(run, "a.txt")
-    text = path.read_text()
-    assert len(text) == count
+    return path.read_text()
+
+
+def test_count(text: str, count: int):
+    assert len(text.splitlines()) == count
+
+
+def test_config(text: str, count: int):
+    assert int(text.split(" ", maxsplit=1)[0]) == count
+
+
+def test_run(text: str, run: Run):
+    line = text.splitlines()[-1]
+    assert line.split(" ", maxsplit=1)[1] == run.info.run_id

{hydraflow-0.6.2 → hydraflow-0.7.2}/tests/run/test_collection.py

@@ -332,44 +332,6 @@ def test_get_param_dict_drop_const(rc: RunCollection):
     assert "r" in params
 
 
-def test_find_dict(rc: RunCollection):
-    run = rc.find({"r": 0})
-    assert run.data.params["p"] == "0"
-
-
-def test_find_kwarg(rc: RunCollection):
-    run = rc.find(r=2)
-    assert run.data.params["p"] == "2"
-
-
-def test_find_none(rc: RunCollection):
-    with pytest.raises(ValueError):
-        rc.find({"r": 10})
-
-
-def test_try_find_none(rc: RunCollection):
-    assert rc.try_find({"r": 10}) is None
-
-
-def test_find_last_dict(rc: RunCollection):
-    run = rc.find_last({"r": 0})
-    assert run.data.params["p"] == "3"
-
-
-def test_find_last_kwarg(rc: RunCollection):
-    run = rc.find_last(r=2)
-    assert run.data.params["p"] == "5"
-
-
-def test_find_last_none(rc: RunCollection):
-    with pytest.raises(ValueError):
-        rc.find_last({"p": 10})
-
-
-def test_try_find_last_none(rc: RunCollection):
-    assert rc.try_find_last({"p": 10}) is None
-
-
 @pytest.mark.parametrize("n_jobs", [0, 1, 2])
 def test_list_runs(rc: RunCollection, n_jobs: int):
     assert len(list_runs(n_jobs=n_jobs)) == 6

hydraflow-0.6.2/tests/context/preemption.py

@@ -1,48 +0,0 @@
-from __future__ import annotations
-
-from dataclasses import dataclass
-from typing import TYPE_CHECKING
-
-import hydra
-from hydra.core.config_store import ConfigStore
-
-import hydraflow
-
-if TYPE_CHECKING:
-    from pathlib import Path
-
-
-@dataclass
-class Config:
-    count: int = 0
-
-
-ConfigStore.instance().store(name="config", node=Config)
-
-
-@hydra.main(version_base=None, config_name="config")
-def app(cfg: Config):
-    hydraflow.set_experiment()
-
-    rc = hydraflow.list_runs()
-
-    if rc.filter(cfg, status="finished", override=True):
-        return
-
-    if run := rc.try_find(cfg, override=True):
-        run_id = run.info.run_id
-    else:
-        run_id = None
-
-    with hydraflow.start_run(cfg, run_id=run_id) as run:
-        log(hydraflow.get_artifact_dir(run))
-
-
-def log(path: Path):
-    file = path / "a.txt"
-    text = file.read_text() if file.exists() else ""
-    file.write_text(text + "a")
-
-
-if __name__ == "__main__":
-    app()