Fix/tests

.github/workflows/ci-oldest-reqs.txt

@@ -1,13 +1,11 @@
-click==7.1.2
-coverage==5.3.1
-filelock==3.0.0
-h5py==2.10.0
-numpy==1.19.0
-packaging==15.0
-pandas==1.0.0
-pymongo==3.10.0
-pytest==6.2.1
-pytest-cov==2.10.1
-pytest-xdist==2.5.0
-tables==3.6.1
-tqdm==4.10.0
+bottleneck == 1.3.4
+freud-analysis == 2.7.0
+coverage[toml] == 6.2
+kneed == 0.8.2
+numpy == 1.21.0
+ruptures == 1.1.5
+scikit-learn == 0.24.0
+pandas == 1.1.3
+hypothesis
+pytest == 6.2.1
+pytest-cov[toml] == 2.12.1

conftest.py

@@ -15,6 +15,12 @@ def seed():
     return seed
 
 
+@pytest.fixture(autouse=True, scope="session")
+def _np_seed():
+    """Set the NumPy random seed for external code."""
+    np.random.seed(3857834)  # noqa: NPY002
+
+
 @pytest.fixture(scope="session")
 def rng():
     """A random number generator for tests that have need for random numbers."""

dupin/__init__.py

@@ -3,3 +3,7 @@
 __author__ = """Brandon Butler"""
 __email__ = "[email protected]"
 __version__ = "0.0.1"
+
+from . import data, detect, errors, postprocessing, preprocessing
+
+__all__ = ("data", "detect", "errors", "postprocessing", "preprocessing")

dupin/data/__init__.py

@@ -59,3 +59,16 @@ def generator():
         pass
 
 """
+from . import aggregate, base, freud, logging, map, reduce, spatial
+from .base import make_generator
+
+__all__ = (
+    "aggregate",
+    "base",
+    "freud",
+    "logging",
+    "map",
+    "reduce",
+    "spatial",
+    "make_generator",
+)

dupin/data/aggregate.py

@@ -5,12 +5,12 @@
 """
 
 
-from typing import Any, Dict, Iterator, Optional, Tuple
+from collections.abc import Iterator
+from typing import Any, Optional
 
 import numpy as np
 
-from dupin import errors
-
+from .. import errors
 from . import base, logging
 
 try:
@@ -58,7 +58,7 @@ def __init__(
         self.logger = logger
 
     def compute(
-        self, iterator: Iterator[Tuple[Tuple[Any, ...], Dict[str, Any]]]
+        self, iterator: Iterator[tuple[tuple[Any, ...], dict[str, Any]]]
     ):
         """Compute signals from generator across the iterator.
 

dupin/data/base.py

@@ -3,7 +3,7 @@
 import typing
 from abc import abstractmethod
 from collections.abc import Callable, Sequence
-from typing import Any, Dict, Optional, Union
+from typing import Any, Optional, Union
 
 import numpy as np
 import numpy.typing as npt
@@ -12,7 +12,7 @@
 GeneratorLike = Union[
     "Generator",
     "DataMap",
-    typing.Callable[..., Dict[str, Union[float, npt.ArrayLike]]],
+    typing.Callable[..., dict[str, Union[float, npt.ArrayLike]]],
 ]
 GeneratorLike.__doc__ = """
 A type hint for objects that act like data generators for dupin.
@@ -213,7 +213,7 @@ class DataReducer(DataModifier):
     """
 
     @abstractmethod
-    def compute(self, distribution: npt.ArrayLike) -> Dict[str, float]:
+    def compute(self, distribution: npt.ArrayLike) -> dict[str, float]:
         """Turn a distribution into scalar features.
 
         Parameters
@@ -284,7 +284,7 @@ class Generator(Callable, PipeComponent):
     @abstractmethod
     def __call__(
         self, *args, **kwargs
-    ) -> Dict[str, Union[float, npt.ArrayLike]]:
+    ) -> dict[str, Union[float, npt.ArrayLike]]:
         """Return the output signal(s) for given inputs.
 
         This method can have an arbitrary signature in subclasses.
@@ -318,8 +318,6 @@ class CustomMap(DataMap):
 
     Parameters
     ----------
-    generator : :py:obj:`~.GeneratorLike`
-        A generator like object to transform.
     custom_function : ``callable`` [`numpy.ndarray`, `dict` ]
         A custom callable that takes in a NumPy array and returns a dictionary
         with keys indicating the tranformation and values the transformed
@@ -333,12 +331,11 @@ class CustomMap(DataMap):
 
     def __init__(
         self,
-        generator: GeneratorLike,
         custom_function: typing.Callable[
-            [npt.ArrayLike], Dict[str, np.ndarray]
+            [npt.ArrayLike], dict[str, np.ndarray]
         ],
     ):
-        super().__init__(generator)
+        super().__init__()
         self.function = custom_function
 
     def compute(self, data: npt.ArrayLike) -> npt.ArrayLike:
@@ -351,8 +348,6 @@ class CustomReducer(DataReducer):
 
     Parameters
     ----------
-    generator: :py:obj:`~.GeneratorLike`
-        A generator like object to reduce.
     custom_function: ``callable`` [`numpy.ndarray`, `dict` [`str`, `float` ]
         A custom callable that takes in a NumPy array and returns a
         dictionary with keys indicating the reduction and values the reduced
@@ -366,7 +361,7 @@ class CustomReducer(DataReducer):
 
     def __init__(
         self,
-        custom_function: typing.Callable[[npt.ArrayLike], Dict[str, float]],
+        custom_function: typing.Callable[[npt.ArrayLike], dict[str, float]],
     ):
         super().__init__()
         self.function = custom_function

dupin/data/freud.py

@@ -13,7 +13,7 @@
 """
 
 from collections.abc import Sequence
-from typing import Any, Dict, List, Union
+from typing import Any, Union
 
 import numpy.typing as npt
 
@@ -70,7 +70,7 @@ class FreudDescriptor(base.Generator):
     def __init__(
         self,
         compute: "freud.util._Compute",
-        attrs: Union[str, List[str], Dict[str, str]],
+        attrs: Union[str, list[str], dict[str, str]],
         compute_method: str = "compute",
     ) -> None:
         if not hasattr(compute, compute_method):
@@ -86,7 +86,7 @@ def __init__(
 
     def __call__(
         self, *args: Any, **kwargs: Any
-    ) -> Dict[str, Union[float, npt.ArrayLike]]:
+    ) -> dict[str, Union[float, npt.ArrayLike]]:
         """Return computed attributes specified in a dictionary.
 
         The keys of the dictionary are the attributes specified, unless a `dict`

dupin/data/logging.py

@@ -23,7 +23,7 @@
 
 import numpy as np
 
-from dupin import errors
+from .. import errors
 
 try:
     import pandas as pd
@@ -71,8 +71,7 @@ def end_frame(self):
             self._current_frame.setdefault(
                 self._context_key, self._current_context
             )
-        if self._current_frame:
-            self._data.append(self._current_frame)
+        self._data.append(self._current_frame)
         self._reset()
 
     def _reset(self):
@@ -104,19 +103,26 @@ def to_dataframe(self):
         -------
             This only works for floating point logged values.
         """
-        # TODO: Extend to other dtypes?
-        if len(self._data) == 0:
+        frame_data = self._first_non_empty(self._data)
+        if frame_data is None:
             return pd.DataFrame()
-        frame_data = self._data[0]
         column_index = pd.MultiIndex.from_tuples(
             _create_column_index(frame_data)
         )
+        # TODO: Extend to other dtypes?
         data_arr = _log_data_to_array(
             self._data,
             np.empty((len(self._data), len(column_index)), dtype=float),
         )
         return pd.DataFrame(data_arr, columns=column_index)
 
+    @staticmethod
+    def _first_non_empty(data: list[dict]):
+        for d in data:
+            if len(d) > 0:
+                return d
+        return None
+
 
 def _create_column_index(log_data):
     """Yield tuples of keys for creating a multi-index from a nested dict."""

dupin/data/map.py

@@ -8,7 +8,7 @@
 map can be found in `dupin.data.spatial.NeighborAveraging`.
 """
 
-from typing import Dict, List, Union
+from typing import Union
 
 import numpy as np
 import numpy.typing as npt
@@ -72,14 +72,17 @@ class Tee(base.DataMap):
 
     def __init__(
         self,
-        maps: List[base.DataMap],
+        maps: list[base.DataMap],
     ):
+        if len(maps) == 0:
+            msg = "Cannot have empty maps sequence."
+            raise ValueError(msg)
         self._maps = maps
         super().__init__()
 
     def compute(
         self, distribution: npt.ArrayLike
-    ) -> Dict[str, Union[float, np.ndarray]]:
+    ) -> dict[str, Union[float, np.ndarray]]:
         """Run all composed map computes."""
         processed_data = {}
         for map_ in self._maps: