hydraflow 0.3.1__tar.gz → 0.3.3__tar.gz

{hydraflow-0.3.1 → hydraflow-0.3.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: hydraflow
-Version: 0.3.1
+Version: 0.3.3
 Summary: Hydraflow integrates Hydra and MLflow to manage and track machine learning experiments.
 Project-URL: Documentation, https://github.com/daizutabi/hydraflow
 Project-URL: Source, https://github.com/daizutabi/hydraflow

{hydraflow-0.3.1 → hydraflow-0.3.3}/mkdocs.yml

@@ -52,5 +52,5 @@ markdown_extensions:
 nav:
   - Home: index.md
   - Usage:
-      - Hydra application: usage/hydra.md
+      - usage/quickstart.md
   - Reference: $api/hydraflow.**

{hydraflow-0.3.1 → hydraflow-0.3.3}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "hydraflow"
-version = "0.3.1"
+version = "0.3.3"
 description = "Hydraflow integrates Hydra and MLflow to manage and track machine learning experiments."
 readme = "README.md"
 license = "MIT"

{hydraflow-0.3.1 → hydraflow-0.3.3}/src/hydraflow/__init__.py

@@ -4,7 +4,13 @@ from .context import chdir_artifact, chdir_hydra, log_run, start_run, watch
 from .mlflow import list_runs, search_runs, set_experiment
 from .progress import multi_tasks_progress, parallel_progress
 from .run_collection import RunCollection
-from .utils import get_artifact_dir, get_hydra_output_dir, load_config
+from .utils import (
+    get_artifact_dir,
+    get_hydra_output_dir,
+    get_overrides,
+    load_config,
+    load_overrides,
+)
 
 __all__ = [
     "RunCollection",
@@ -12,8 +18,10 @@ __all__ = [
     "chdir_hydra",
     "get_artifact_dir",
     "get_hydra_output_dir",
+    "get_overrides",
     "list_runs",
     "load_config",
+    "load_overrides",
     "log_run",
     "multi_tasks_progress",
     "parallel_progress",

{hydraflow-0.3.1 → hydraflow-0.3.3}/src/hydraflow/param.py

@@ -72,4 +72,4 @@ def _match_tuple(param: str, value: tuple) -> bool | None:
     if type(value[0]) is not type(value[1]):
         return None
 
-    return value[0] <= type(value[0])(param) < value[1]  # type: ignore
+    return value[0] <= type(value[0])(param) <= value[1]  # type: ignore

{hydraflow-0.3.1 → hydraflow-0.3.3}/src/hydraflow/run_collection.py

@@ -24,12 +24,12 @@ from itertools import chain
 from typing import TYPE_CHECKING, Any, Concatenate, ParamSpec, TypeVar, overload
 
 from mlflow.entities import RunStatus
-from polars.dataframe import DataFrame
 
 import hydraflow.param
-from hydraflow.config import collect_params, iter_params
+from hydraflow.config import iter_params
 from hydraflow.run_data import RunCollectionData
 from hydraflow.run_info import RunCollectionInfo
+from hydraflow.utils import load_config
 
 if TYPE_CHECKING:
     from collections.abc import Callable, Iterator
@@ -239,8 +239,8 @@ class RunCollection:
         The filtering supports:
         - Exact matches for single values.
         - Membership checks for lists of values.
-        - Range checks for tuples of two values (inclusive of the lower bound
-          and exclusive of the upper bound).
+        - Range checks for tuples of two values (inclusive of both the lower
+          and upper bound).
 
         Args:
             config (object | None): The configuration object to filter the runs.
@@ -476,7 +476,7 @@ class RunCollection:
         """
         return (func(run, *args, **kwargs) for run in self)
 
-    def map_run_id(
+    def map_id(
         self,
         func: Callable[Concatenate[str, P], T],
         *args: P.args,
@@ -516,7 +516,7 @@ class RunCollection:
             in the collection.
 
         """
-        return (func(config, *args, **kwargs) for config in self.data.config)
+        return (func(load_config(run), *args, **kwargs) for run in self)
 
     def map_uri(
         self,
@@ -569,8 +569,8 @@ class RunCollection:
 
     def group_by(
         self,
-        *names: str | list[str],
-    ) -> dict[tuple[str | None, ...], RunCollection]:
+        names: str | list[str],
+    ) -> dict[str | None | tuple[str | None, ...], RunCollection]:
         """Group runs by specified parameter names.
 
         Group the runs in the collection based on the values of the
@@ -578,33 +578,27 @@ class RunCollection:
         form a key in the returned dictionary.
 
         Args:
-            *names (str | list[str]): The names of the parameters to group by.
+            names (str | list[str]): The names of the parameters to group by.
                 This can be a single parameter name or multiple names provided
                 as separate arguments or as a list.
 
         Returns:
-            dict[tuple[str | None, ...], RunCollection]: A dictionary where the keys
-            are tuples of parameter values and the values are RunCollection objects
-            containing the runs that match those parameter values.
+            dict[str | None | tuple[str | None, ...], RunCollection]: A
+            dictionary where the keys are tuples of parameter values and the
+            values are `RunCollection` objects containing the runs that match
+            those parameter values.
 
         """
-        grouped_runs: dict[tuple[str | None, ...], list[Run]] = {}
+        grouped_runs: dict[str | None | tuple[str | None, ...], list[Run]] = {}
+        is_list = isinstance(names, list)
         for run in self._runs:
-            key = get_params(run, *names)
+            key = get_params(run, names)
+            if not is_list:
+                key = key[0]
             grouped_runs.setdefault(key, []).append(run)
 
         return {key: RunCollection(runs) for key, runs in grouped_runs.items()}
 
-    @property
-    def config(self) -> DataFrame:
-        """Get the runs' configurations as a polars DataFrame.
-
-        Returns:
-            A polars DataFrame containing the runs' configurations.
-
-        """
-        return DataFrame(self.map_config(collect_params))
-
 
 def _param_matches(run: Run, key: str, value: Any) -> bool:
     params = run.data.params
@@ -637,8 +631,8 @@ def filter_runs(
     The filtering supports:
     - Exact matches for single values.
     - Membership checks for lists of values.
-    - Range checks for tuples of two values (inclusive of the lower bound and
-      exclusive of the upper bound).
+    - Range checks for tuples of two values (inclusive of both the lower and
+      upper bound).
 
     Args:
         runs (list[Run]): The list of runs to filter.

hydraflow-0.3.3/src/hydraflow/run_data.py

@@ -0,0 +1,57 @@
+"""Provide data about `RunCollection` instances."""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+from polars.dataframe import DataFrame
+
+from hydraflow.config import collect_params
+
+if TYPE_CHECKING:
+    from collections.abc import Iterable
+    from typing import Any
+
+    from hydraflow.run_collection import RunCollection
+
+
+class RunCollectionData:
+    """Provide data about a `RunCollection` instance."""
+
+    def __init__(self, runs: RunCollection) -> None:
+        self._runs = runs
+
+    @property
+    def params(self) -> dict[str, list[str]]:
+        """Get the parameters for each run in the collection."""
+        return _to_dict(run.data.params for run in self._runs)
+
+    @property
+    def metrics(self) -> dict[str, list[float]]:
+        """Get the metrics for each run in the collection."""
+        return _to_dict(run.data.metrics for run in self._runs)
+
+    @property
+    def config(self) -> DataFrame:
+        """Get the runs' configurations as a polars DataFrame.
+
+        Returns:
+            A polars DataFrame containing the runs' configurations.
+
+        """
+        return DataFrame(self._runs.map_config(collect_params))
+
+
+def _to_dict(it: Iterable[dict[str, Any]]) -> dict[str, list[Any]]:
+    """Convert an iterable of dictionaries to a dictionary of lists."""
+    data = list(it)
+    if not data:
+        return {}
+
+    keys = []
+    for d in data:
+        for key in d:
+            if key not in keys:
+                keys.append(key)
+
+    return {key: [x.get(key) for x in data] for key in keys}

{hydraflow-0.3.1 → hydraflow-0.3.3}/src/hydraflow/utils.py

@@ -68,6 +68,11 @@ def get_hydra_output_dir(run: Run | None = None) -> Path:
     raise FileNotFoundError
 
 
+def get_overrides() -> list[str]:
+    """Retrieve the overrides for the current run."""
+    return HydraConfig.get().overrides.task
+
+
 def load_config(run: Run) -> DictConfig:
     """Load the configuration for a given run.
 
@@ -86,3 +91,23 @@ def load_config(run: Run) -> DictConfig:
     """
     path = get_artifact_dir(run) / ".hydra/config.yaml"
     return OmegaConf.load(path)  # type: ignore
+
+
+def load_overrides(run: Run) -> list[str]:
+    """Load the overrides for a given run.
+
+    This function loads the overrides for the provided Run instance
+    by downloading the overrides file from the MLflow artifacts and
+    loading it using OmegaConf. It returns an empty config if
+    `.hydra/overrides.yaml` is not found in the run's artifact directory.
+
+    Args:
+        run (Run): The Run instance for which to load the overrides.
+
+    Returns:
+        The loaded overrides as a list of strings. Returns an empty list
+        if the overrides file is not found.
+
+    """
+    path = get_artifact_dir(run) / ".hydra/overrides.yaml"
+    return [str(x) for x in OmegaConf.load(path)]

{hydraflow-0.3.1 → hydraflow-0.3.3}/tests/scripts/app.py

@@ -31,7 +31,7 @@ def app(cfg: MySQLConfig):
         Path("chdir_hydra.txt").write_text(path.as_posix())
 
     hydraflow.set_experiment(prefix="_", suffix="_")
-    with hydraflow.start_run(cfg):
+    with hydraflow.start_run(cfg) as run:
         log.info(f"START, {cfg.host}, {cfg.port} ")
 
         artifact_dir = hydraflow.get_artifact_dir()
@@ -50,6 +50,8 @@ def app(cfg: MySQLConfig):
         if cfg.host == "x":
             mlflow.log_metric("m", cfg.port + 10, 2)
 
+        assert hydraflow.get_overrides() == hydraflow.load_overrides(run)
+
         log.info("END")
 
 

{hydraflow-0.3.1 → hydraflow-0.3.3}/tests/test_app.py

@@ -8,7 +8,7 @@ from typing import TYPE_CHECKING
 import mlflow
 import pytest
 from mlflow.entities import RunStatus
-from omegaconf import ListConfig, OmegaConf
+from omegaconf import OmegaConf
 
 if TYPE_CHECKING:
     from omegaconf import DictConfig
@@ -92,33 +92,30 @@ def test_app_info_run_id(rc: RunCollection):
 
 def test_app_data_params(rc: RunCollection):
     params = rc.data.params
-    assert params[0] == {"port": "1", "host": "x", "values": "[1, 2, 3]"}
-    assert params[1] == {"port": "2", "host": "x", "values": "[1, 2, 3]"}
-    assert params[2] == {"port": "1", "host": "y", "values": "[1, 2, 3]"}
-    assert params[3] == {"port": "2", "host": "y", "values": "[1, 2, 3]"}
+    assert params["port"] == ["1", "2", "1", "2"]
+    assert params["host"] == ["x", "x", "y", "y"]
+    assert params["values"] == ["[1, 2, 3]", "[1, 2, 3]", "[1, 2, 3]", "[1, 2, 3]"]
 
 
 def test_app_data_metrics(rc: RunCollection):
     metrics = rc.data.metrics
-    assert metrics[0] == {"m": 11, "watch": 3}
-    assert metrics[1] == {"m": 12, "watch": 3}
-    assert metrics[2] == {"m": 2, "watch": 3}
-    assert metrics[3] == {"m": 3, "watch": 3}
+    assert metrics["m"] == [11, 12, 2, 3]
+    assert metrics["watch"] == [3, 3, 3, 3]
 
 
 def test_app_data_config(rc: RunCollection):
     config = rc.data.config
-    assert config[0].port == 1
-    assert config[1].port == 2
-    assert config[2].host == "y"
-    assert config[3].host == "y"
+    assert config["port"].to_list() == [1, 2, 1, 2]
+    assert config["host"].to_list() == ["x", "x", "y", "y"]
 
 
 def test_app_data_config_list(rc: RunCollection):
     config = rc.data.config
-    assert isinstance(config[0]["values"], ListConfig)
-    assert not isinstance(config[0]["values"], list)
-    assert config[0]["values"] == [1, 2, 3]
+    values = config["values"].to_list()
+    assert str(config.select("values").dtypes) == "[List(Int64)]"
+    for x in values:
+        assert isinstance(x, list)
+        assert x == [1, 2, 3]
 
 
 def test_app_info_artifact_uri(rc: RunCollection):
@@ -160,14 +157,19 @@ def test_app_map_config(rc: RunCollection):
 def test_app_group_by(rc: RunCollection):
     grouped = rc.group_by("host")
     assert len(grouped) == 2
-    x = {"port": "1", "host": "x", "values": "[1, 2, 3]"}
-    assert grouped[("x",)].data.params[0] == x
-    x = {"port": "2", "host": "x", "values": "[1, 2, 3]"}
-    assert grouped[("x",)].data.params[1] == x
-    x = {"port": "1", "host": "y", "values": "[1, 2, 3]"}
-    assert grouped[("y",)].data.params[0] == x
-    x = {"port": "2", "host": "y", "values": "[1, 2, 3]"}
-    assert grouped[("y",)].data.params[1] == x
+    assert grouped["x"].data.params["port"] == ["1", "2"]
+    assert grouped["x"].data.params["host"] == ["x", "x"]
+    assert grouped["x"].data.params["values"] == ["[1, 2, 3]", "[1, 2, 3]"]
+    assert grouped["y"].data.params["port"] == ["1", "2"]
+    assert grouped["y"].data.params["host"] == ["y", "y"]
+    assert grouped["y"].data.params["values"] == ["[1, 2, 3]", "[1, 2, 3]"]
+
+
+def test_app_group_by_list(rc: RunCollection):
+    grouped = rc.group_by(["host"])
+    assert len(grouped) == 2
+    assert ("x",) in grouped
+    assert ("y",) in grouped
 
 
 def test_app_filter_list(rc: RunCollection):
@@ -177,18 +179,3 @@ def test_app_filter_list(rc: RunCollection):
     assert len(filtered) == 4
     filtered = rc.filter(values=[1])
     assert not filtered
-
-
-def test_config(rc: RunCollection):
-    df = rc.config
-    assert df.columns == ["host", "port", "values"]
-    assert df.shape == (4, 3)
-    assert df.select("host").to_series().to_list() == ["x", "x", "y", "y"]
-    assert df.select("port").to_series().to_list() == [1, 2, 1, 2]
-    assert str(df.select("values").dtypes) == "[List(Int64)]"
-    assert df.select("values").to_series().to_list() == [
-        [1, 2, 3],
-        [1, 2, 3],
-        [1, 2, 3],
-        [1, 2, 3],
-    ]

{hydraflow-0.3.1 → hydraflow-0.3.3}/tests/test_log_run.py

@@ -50,7 +50,7 @@ def read_log(run_id: str, path: str) -> str:
 
 
 def test_load_config(run: Run):
-    from hydraflow.run_data import load_config
+    from hydraflow.utils import load_config
 
     log = read_log(run.info.run_id, "log_run.log")
     assert "START" in log
@@ -63,6 +63,18 @@ def test_load_config(run: Run):
     assert cfg.port == int(port)
 
 
+def test_load_overrides(run: Run):
+    from hydraflow.utils import load_overrides
+
+    log = read_log(run.info.run_id, "log_run.log")
+    assert "START" in log
+    assert "END" in log
+
+    host, port = log.splitlines()[0].split("START,")[-1].split(",")
+
+    assert load_overrides(run) == [f"host={host.strip()}", f"port={port.strip()}"]
+
+
 def test_info(run: Run):
     log = read_log(run.info.run_id, "artifact_dir.txt")
     a, b = log.split(" ")

{hydraflow-0.3.1 → hydraflow-0.3.3}/tests/test_run_collection.py

@@ -28,12 +28,12 @@ def rc(monkeypatch, tmp_path):
     return x
 
 
-def test_run_collection_bool_false():
+def test_bool_false():
     assert not RunCollection([])
     assert bool(RunCollection.from_list([])) is False
 
 
-def test_run_collection_bool_true(rc: RunCollection):
+def test_bool_true(rc: RunCollection):
     assert rc
     assert bool(rc) is True
 
@@ -90,7 +90,7 @@ def test_filter_tuple(run_list: list[Run]):
     from hydraflow.run_collection import filter_runs
 
     x = filter_runs(run_list, p=(1, 3))
-    assert len(x) == 2
+    assert len(x) == 3
 
 
 def test_filter_invalid_param(run_list: list[Run]):
@@ -139,45 +139,45 @@ def test_chdir_artifact_list(i: int, run_list: list[Run]):
     assert not Path("abc.txt").exists()
 
 
-def test_runs_repr(rc: RunCollection):
+def test_repr(rc: RunCollection):
     assert repr(rc) == "RunCollection(6)"
 
 
-def test_runs_first(rc: RunCollection):
+def test_first(rc: RunCollection):
     run = rc.first()
     assert isinstance(run, Run)
     assert run.data.params["p"] == "0"
 
 
-def test_runs_first_empty(rc: RunCollection):
+def test_first_empty(rc: RunCollection):
     rc._runs = []
     with pytest.raises(ValueError):
         rc.first()
 
 
-def test_runs_try_first_none(rc: RunCollection):
+def test_try_first_none(rc: RunCollection):
     rc._runs = []
     assert rc.try_first() is None
 
 
-def test_runs_last(rc: RunCollection):
+def test_last(rc: RunCollection):
     run = rc.last()
     assert isinstance(run, Run)
     assert run.data.params["p"] == "5"
 
 
-def test_runs_last_empty(rc: RunCollection):
+def test_last_empty(rc: RunCollection):
     rc._runs = []
     with pytest.raises(ValueError):
         rc.last()
 
 
-def test_runs_try_last_none(rc: RunCollection):
+def test_try_last_none(rc: RunCollection):
     rc._runs = []
     assert rc.try_last() is None
 
 
-def test_runs_filter(rc: RunCollection):
+def test_filter(rc: RunCollection):
     assert len(rc.filter()) == 6
     assert len(rc.filter({})) == 6
     assert len(rc.filter({"p": 1})) == 1
@@ -192,14 +192,14 @@ def test_runs_filter(rc: RunCollection):
     assert len(rc.filter(r=0)) == 2
 
 
-def test_runs_get(rc: RunCollection):
+def test_get(rc: RunCollection):
     run = rc.get({"p": 4})
     assert isinstance(run, Run)
     run = rc.get(p=2)
     assert isinstance(run, Run)
 
 
-def test_runs_try_get(rc: RunCollection):
+def test_try_get(rc: RunCollection):
     run = rc.try_get({"p": 5})
     assert isinstance(run, Run)
     run = rc.try_get(p=1)
@@ -208,7 +208,7 @@ def test_runs_try_get(rc: RunCollection):
     assert run is None
 
 
-def test_runs_get_params_names(rc: RunCollection):
+def test_get_param_names(rc: RunCollection):
     names = rc.get_param_names()
     assert len(names) == 3
     assert "p" in names
@@ -216,14 +216,14 @@ def test_runs_get_params_names(rc: RunCollection):
     assert "r" in names
 
 
-def test_runs_get_params_dict(rc: RunCollection):
+def test_get_param_dict(rc: RunCollection):
     params = rc.get_param_dict()
     assert params["p"] == ["0", "1", "2", "3", "4", "5"]
     assert params["q"] == ["0", "None"]
     assert params["r"] == ["0", "1", "2"]
 
 
-def test_runs_get_params_dict_drop_const(rc: RunCollection):
+def test_get_param_dict_drop_const(rc: RunCollection):
     rc_ = rc.filter(q=0)
     params = rc_.get_param_dict(drop_const=True)
     assert len(params) == 2
@@ -232,7 +232,7 @@ def test_runs_get_params_dict_drop_const(rc: RunCollection):
     assert "r" in params
 
 
-def test_runs_find(rc: RunCollection):
+def test_find(rc: RunCollection):
     run = rc.find({"r": 0})
     assert isinstance(run, Run)
     assert run.data.params["p"] == "0"
@@ -241,17 +241,17 @@ def test_runs_find(rc: RunCollection):
     assert run.data.params["p"] == "2"
 
 
-def test_runs_find_none(rc: RunCollection):
+def test_find_none(rc: RunCollection):
     with pytest.raises(ValueError):
         rc.find({"r": 10})
 
 
-def test_runs_try_find_none(rc: RunCollection):
+def test_try_find_none(rc: RunCollection):
     run = rc.try_find({"r": 10})
     assert run is None
 
 
-def test_runs_find_last(rc: RunCollection):
+def test_find_last(rc: RunCollection):
     run = rc.find_last({"r": 0})
     assert isinstance(run, Run)
     assert run.data.params["p"] == "3"
@@ -260,12 +260,12 @@ def test_runs_find_last(rc: RunCollection):
     assert run.data.params["p"] == "5"
 
 
-def test_runs_find_last_none(rc: RunCollection):
+def test_find_last_none(rc: RunCollection):
     with pytest.raises(ValueError):
         rc.find_last({"p": 10})
 
 
-def test_runs_try_find_last_none(rc: RunCollection):
+def test_try_find_last_none(rc: RunCollection):
     run = rc.try_find_last({"p": 10})
     assert run is None
 
@@ -313,42 +313,42 @@ def test_list_runs_none(rc, runs2):
     assert not no_runs
 
 
-def test_run_collection_map(rc: RunCollection):
+def test_map(rc: RunCollection):
     results = list(rc.map(lambda run: run.info.run_id))
     assert len(results) == len(rc._runs)
     assert all(isinstance(run_id, str) for run_id in results)
 
 
-def test_run_collection_map_args(rc: RunCollection):
+def test_map_args(rc: RunCollection):
     results = list(rc.map(lambda run, x: run.info.run_id + x, "test"))
     assert all(x.endswith("test") for x in results)
 
 
-def test_run_collection_map_run_id(rc: RunCollection):
-    results = list(rc.map_run_id(lambda run_id: run_id))
+def test_map_id(rc: RunCollection):
+    results = list(rc.map_id(lambda run_id: run_id))
     assert len(results) == len(rc._runs)
     assert all(isinstance(run_id, str) for run_id in results)
 
 
-def test_run_collection_map_run_id_kwargs(rc: RunCollection):
-    results = list(rc.map_run_id(lambda run_id, x: x + run_id, x="test"))
+def test_map_id_kwargs(rc: RunCollection):
+    results = list(rc.map_id(lambda run_id, x: x + run_id, x="test"))
     assert all(x.startswith("test") for x in results)
 
 
-def test_run_collection_map_uri(rc: RunCollection):
+def test_map_uri(rc: RunCollection):
     results = list(rc.map_uri(lambda uri: uri))
     assert len(results) == len(rc._runs)
     assert all(isinstance(uri, str | type(None)) for uri in results)
 
 
-def test_run_collection_map_dir(rc: RunCollection):
+def test_map_dir(rc: RunCollection):
     results = list(rc.map_dir(lambda dir_path, x: dir_path / x, "a.csv"))
     assert len(results) == len(rc._runs)
     assert all(isinstance(dir_path, Path) for dir_path in results)
     assert all(dir_path.stem == "a" for dir_path in results)
 
 
-def test_run_collection_sort(rc: RunCollection):
+def test_sort(rc: RunCollection):
     rc.sort(key=lambda x: x.data.params["p"])
     assert [run.data.params["p"] for run in rc] == ["0", "1", "2", "3", "4", "5"]
 
@@ -356,7 +356,7 @@ def test_run_collection_sort(rc: RunCollection):
     assert [run.data.params["p"] for run in rc] == ["5", "4", "3", "2", "1", "0"]
 
 
-def test_run_collection_iter(rc: RunCollection):
+def test_iter(rc: RunCollection):
     assert list(rc) == rc._runs
 
 
@@ -366,39 +366,39 @@ def test_run_collection_getitem(rc: RunCollection, i: int):
 
 
 @pytest.mark.parametrize("i", range(6))
-def test_run_collection_getitem_slice(rc: RunCollection, i: int):
+def test_getitem_slice(rc: RunCollection, i: int):
     assert rc[i : i + 2]._runs == rc._runs[i : i + 2]
 
 
 @pytest.mark.parametrize("i", range(6))
-def test_run_collection_getitem_slice_step(rc: RunCollection, i: int):
+def test_getitem_slice_step(rc: RunCollection, i: int):
     assert rc[i::2]._runs == rc._runs[i::2]
 
 
 @pytest.mark.parametrize("i", range(6))
-def test_run_collection_getitem_slice_step_neg(rc: RunCollection, i: int):
+def test_getitem_slice_step_neg(rc: RunCollection, i: int):
     assert rc[i::-2]._runs == rc._runs[i::-2]
 
 
-def test_run_collection_take(rc: RunCollection):
+def test_take(rc: RunCollection):
     assert rc.take(3)._runs == rc._runs[:3]
     assert len(rc.take(4)) == 4
     assert rc.take(10)._runs == rc._runs
 
 
-def test_run_collection_take_neg(rc: RunCollection):
+def test_take_neg(rc: RunCollection):
     assert rc.take(-3)._runs == rc._runs[-3:]
     assert len(rc.take(-4)) == 4
     assert rc.take(-10)._runs == rc._runs
 
 
 @pytest.mark.parametrize("i", range(6))
-def test_run_collection_contains(rc: RunCollection, i: int):
+def test_contains(rc: RunCollection, i: int):
     assert rc[i] in rc
     assert rc._runs[i] in rc
 
 
-def test_run_collection_group_by(rc: RunCollection):
+def test_group_by(rc: RunCollection):
     grouped = rc.group_by(["p"])
     assert len(grouped) == 6
     assert all(isinstance(group, RunCollection) for group in grouped.values())

{hydraflow-0.3.1 → hydraflow-0.3.3}/tests/test_run_data.py

@@ -26,18 +26,17 @@ def runs(monkeypatch, tmp_path):
 
 
 def test_data_params(runs: RunCollection):
-    assert runs.data.params == [{"p": "0"}, {"p": "1"}, {"p": "2"}]
+    assert runs.data.params["p"] == ["0", "1", "2"]
 
 
 def test_data_metrics(runs: RunCollection):
     m = runs.data.metrics
-    assert m[0] == {"metric1": 1, "metric2": 2}
-    assert m[1] == {"metric1": 2, "metric2": 3}
-    assert m[2] == {"metric1": 3, "metric2": 4}
+    assert m["metric1"] == [1, 2, 3]
+    assert m["metric2"] == [2, 3, 4]
 
 
 def test_data_empty_run_collection():
     rc = RunCollection([])
-    assert rc.data.params == []
-    assert rc.data.metrics == []
-    assert rc.data.config == []
+    assert rc.data.params == {}
+    assert rc.data.metrics == {}
+    assert len(rc.data.config) == 0

hydraflow-0.3.1/src/hydraflow/run_data.py

@@ -1,34 +0,0 @@
-"""Provide data about `RunCollection` instances."""
-
-from __future__ import annotations
-
-from typing import TYPE_CHECKING
-
-from hydraflow.utils import load_config
-
-if TYPE_CHECKING:
-    from omegaconf import DictConfig
-
-    from hydraflow.run_collection import RunCollection
-
-
-class RunCollectionData:
-    """Provide data about a `RunCollection` instance."""
-
-    def __init__(self, runs: RunCollection) -> None:
-        self._runs = runs
-
-    @property
-    def params(self) -> list[dict[str, str]]:
-        """Get the parameters for each run in the collection."""
-        return [run.data.params for run in self._runs]
-
-    @property
-    def metrics(self) -> list[dict[str, float]]:
-        """Get the metrics for each run in the collection."""
-        return [run.data.metrics for run in self._runs]
-
-    @property
-    def config(self) -> list[DictConfig]:
-        """Get the configuration for each run in the collection."""
-        return [load_config(run) for run in self._runs]