bump min deps for 0.15

ci/requirements/py36-bare-minimum.yml

@@ -7,5 +7,5 @@ dependencies:
   - pytest
   - pytest-cov
   - pytest-env
-  - numpy=1.14
-  - pandas=0.24
+  - numpy=1.15
+  - pandas=0.25

ci/requirements/py36-min-all-deps.yml

@@ -29,8 +29,8 @@ dependencies:
   - nc-time-axis=1.2
   - netcdf4=1.4
   - numba=0.44
-  - numpy=1.14
-  - pandas=0.24
+  - numpy=1.15
+  - pandas=0.25
   # - pint  # See py36-min-nep18.yml
   - pip
   - pseudonetcdf=3.0
@@ -40,7 +40,7 @@ dependencies:
   - pytest-cov
   - pytest-env
   - rasterio=1.0
-  - scipy=1.0  # Policy allows for 1.2, but scipy>=1.1 breaks numpy=1.14
+  - scipy=1.3
   - seaborn=0.9
   # - sparse  # See py36-min-nep18.yml
   - toolz=0.10

ci/requirements/py36-min-nep18.yml

@@ -9,7 +9,7 @@ dependencies:
   - dask=2.4
   - distributed=2.4
   - numpy=1.17
-  - pandas=0.24
+  - pandas=0.25
   - pint=0.9  # Actually not enough as it doesn't implement __array_function__yet!
   - pytest
   - pytest-cov

doc/installing.rst

@@ -7,8 +7,8 @@ Required dependencies
 ---------------------
 
 - Python (3.6 or later)
-- `numpy <http://www.numpy.org/>`__ (1.14 or later)
-- `pandas <http://pandas.pydata.org/>`__ (0.24 or later)
+- `numpy <http://www.numpy.org/>`__ (1.15 or later)
+- `pandas <http://pandas.pydata.org/>`__ (0.25 or later)
 
 Optional dependencies
 ---------------------

doc/whats-new.rst

@@ -21,7 +21,13 @@ v0.15.0 (unreleased)
 
 Breaking changes
 ~~~~~~~~~~~~~~~~
-- Bumped minimum ``dask`` version to 2.2.
+- Bumped minimum tested versions for dependencies:
+  - numpy 1.15
+  - pandas 0.25
+  - dask 2.2
+  - distributed 2.2
+  - scipy 1.3
+
 - Remove ``compat`` and ``encoding`` kwargs from ``DataArray``, which
   have been deprecated since 0.12. (:pull:`3650`). 
   Instead, specify the encoding when writing to disk or set 

setup.cfg

@@ -79,8 +79,8 @@ zip_safe = True
 include_package_data = True
 python_requires = >=3.6
 install_requires =
-    numpy >= 1.14
-    pandas >= 0.24
+    numpy >= 1.15
+    pandas >= 0.25
 setup_requires = setuptools_scm
 
 [options.package_data]

xarray/core/dataarray.py

@@ -2969,7 +2969,7 @@ def quantile(
 
         See Also
         --------
-        numpy.nanpercentile, pandas.Series.quantile, Dataset.quantile
+        numpy.nanquantile, pandas.Series.quantile, Dataset.quantile
 
         Examples
         --------

xarray/core/dataset.py

@@ -4466,12 +4466,7 @@ def _set_sparse_data_from_dataframe(
 
         idx = dataframe.index
         if isinstance(idx, pd.MultiIndex):
-            try:
-                codes = idx.codes
-            except AttributeError:
-                # deprecated since pandas 0.24
-                codes = idx.labels
-            coords = np.stack([np.asarray(code) for code in codes], axis=0)
+            coords = np.stack([np.asarray(code) for code in idx.codes], axis=0)
             is_sorted = idx.is_lexsorted
         else:
             coords = np.arange(idx.size).reshape(1, -1)
@@ -5171,7 +5166,7 @@ def quantile(
 
         See Also
         --------
-        numpy.nanpercentile, pandas.Series.quantile, DataArray.quantile
+        numpy.nanquantile, pandas.Series.quantile, DataArray.quantile
 
         Examples
         --------

xarray/core/groupby.py

@@ -595,7 +595,7 @@ def quantile(self, q, dim=None, interpolation="linear", keep_attrs=None):
 
         See Also
         --------
-        numpy.nanpercentile, pandas.Series.quantile, Dataset.quantile,
+        numpy.nanquantile, pandas.Series.quantile, Dataset.quantile,
         DataArray.quantile
 
         Examples

xarray/core/nputils.py

@@ -3,6 +3,8 @@
 import numpy as np
 import pandas as pd
 
+from numpy.core.multiarray import normalize_axis_index
+
 try:
     import bottleneck as bn
 
@@ -13,29 +15,20 @@
     _USE_BOTTLENECK = False
 
 
-def _validate_axis(data, axis):
-    ndim = data.ndim
-    if not -ndim <= axis < ndim:
-        raise IndexError(f"axis {axis!r} out of bounds [-{ndim}, {ndim})")
-    if axis < 0:
-        axis += ndim
-    return axis
-
-
 def _select_along_axis(values, idx, axis):
     other_ind = np.ix_(*[np.arange(s) for s in idx.shape])
     sl = other_ind[:axis] + (idx,) + other_ind[axis:]
     return values[sl]
 
 
 def nanfirst(values, axis):
-    axis = _validate_axis(values, axis)
+    axis = normalize_axis_index(axis, values.ndim)
     idx_first = np.argmax(~pd.isnull(values), axis=axis)
     return _select_along_axis(values, idx_first, axis)
 
 
 def nanlast(values, axis):
-    axis = _validate_axis(values, axis)
+    axis = normalize_axis_index(axis, values.ndim)
     rev = (slice(None),) * axis + (slice(None, None, -1),)
     idx_last = -1 - np.argmax(~pd.isnull(values)[rev], axis=axis)
     return _select_along_axis(values, idx_last, axis)
@@ -186,7 +179,7 @@ def _rolling_window(a, window, axis=-1):
     This function is taken from https://github.com/numpy/numpy/pull/31
     but slightly modified to accept axis option.
     """
-    axis = _validate_axis(a, axis)
+    axis = normalize_axis_index(axis, a.ndim)
     a = np.swapaxes(a, axis, -1)
 
     if window < 1:

xarray/core/variable.py

@@ -1722,7 +1722,7 @@ def quantile(self, q, dim=None, interpolation="linear", keep_attrs=None):
 
         See Also
         --------
-        numpy.nanpercentile, pandas.Series.quantile, Dataset.quantile,
+        numpy.nanquantile, pandas.Series.quantile, Dataset.quantile,
         DataArray.quantile
         """
 
@@ -1734,10 +1734,6 @@ def quantile(self, q, dim=None, interpolation="linear", keep_attrs=None):
         scalar = utils.is_scalar(q)
         q = np.atleast_1d(np.asarray(q, dtype=np.float64))
 
-        # TODO: remove once numpy >= 1.15.0 is the minimum requirement
-        if np.count_nonzero(q < 0.0) or np.count_nonzero(q > 1.0):
-            raise ValueError("Quantiles must be in the range [0, 1]")
-
         if dim is None:
             dim = self.dims
 
@@ -1746,9 +1742,7 @@ def quantile(self, q, dim=None, interpolation="linear", keep_attrs=None):
 
         def _wrapper(npa, **kwargs):
             # move quantile axis to end. required for apply_ufunc
-
-            # TODO: use np.nanquantile once numpy >= 1.15.0 is the minimum requirement
-            return np.moveaxis(np.nanpercentile(npa, **kwargs), 0, -1)
+            return np.moveaxis(np.nanquantile(npa, **kwargs), 0, -1)
 
         axis = np.arange(-1, -1 * len(dim) - 1, -1)
         result = apply_ufunc(
@@ -1760,7 +1754,7 @@ def _wrapper(npa, **kwargs):
             output_dtypes=[np.float64],
             output_sizes={"quantile": len(q)},
             dask="parallelized",
-            kwargs={"q": q * 100, "axis": axis, "interpolation": interpolation},
+            kwargs={"q": q, "axis": axis, "interpolation": interpolation},
         )
 
         # for backward compatibility

xarray/plot/utils.py

@@ -547,10 +547,6 @@ def _add_colorbar(primitive, ax, cbar_ax, cbar_kwargs, cmap_params):
 
 def _rescale_imshow_rgb(darray, vmin, vmax, robust):
     assert robust or vmin is not None or vmax is not None
-    # TODO: remove when min numpy version is bumped to 1.13
-    # There's a cyclic dependency via DataArray, so we can't import from
-    # xarray.ufuncs in global scope.
-    from xarray.ufuncs import maximum, minimum
 
     # Calculate vmin and vmax automatically for `robust=True`
     if robust:
@@ -579,9 +575,7 @@ def _rescale_imshow_rgb(darray, vmin, vmax, robust):
     # After scaling, downcast to 32-bit float.  This substantially reduces
     # memory usage after we hand `darray` off to matplotlib.
     darray = ((darray.astype("f8") - vmin) / (vmax - vmin)).astype("f4")
-    with warnings.catch_warnings():
-        warnings.filterwarnings("ignore", "xarray.ufuncs", PendingDeprecationWarning)
-        return minimum(maximum(darray, 0), 1)
+    return np.minimum(np.maximum(darray, 0), 1)
 
 
 def _update_axes(

xarray/tests/test_dataarray.py

@@ -80,7 +80,7 @@ def test_repr_multiindex(self):
         assert expected == repr(self.mda)
 
     @pytest.mark.skipif(
-        LooseVersion(np.__version__) < "1.15",
+        LooseVersion(np.__version__) < "1.16",
         reason="old versions of numpy have different printing behavior",
     )
     def test_repr_multiindex_long(self):

xarray/tests/test_units.py

@@ -1838,7 +1838,10 @@ def test_squeeze(self, dtype):
         "func",
         (
             method("coarsen", windows={"y": 2}, func=np.mean),
-            method("quantile", q=[0.25, 0.75]),
+            pytest.param(
+                method("quantile", q=[0.25, 0.75]),
+                marks=pytest.mark.xfail(reason="nanquantile not implemented"),
+            ),
             pytest.param(
                 method("rank", dim="x"),
                 marks=pytest.mark.xfail(reason="rank not implemented for non-ndarray"),
@@ -3401,7 +3404,10 @@ def test_stacking_reordering(self, func, dtype):
             method("diff", dim="x"),
             method("differentiate", coord="x"),
             method("integrate", dim="x"),
-            method("quantile", q=[0.25, 0.75]),
+            pytest.param(
+                method("quantile", q=[0.25, 0.75]),
+                marks=pytest.mark.xfail(reason="nanquantile not implemented"),
+            ),
             method("reduce", func=np.sum, dim="x"),
             pytest.param(
                 lambda x: x.dot(x),
@@ -3491,7 +3497,10 @@ def test_resample(self, dtype):
             method("assign_coords", z=(["x"], np.arange(5) * unit_registry.s)),
             method("first"),
             method("last"),
-            method("quantile", q=np.array([0.25, 0.5, 0.75]), dim="x"),
+            pytest.param(
+                method("quantile", q=[0.25, 0.5, 0.75], dim="x"),
+                marks=pytest.mark.xfail(reason="nanquantile not implemented"),
+            ),
         ),
         ids=repr,
     )
@@ -4929,7 +4938,10 @@ def test_reindex_like(self, unit, error, dtype):
             method("diff", dim="x"),
             method("differentiate", coord="x"),
             method("integrate", coord="x"),
-            method("quantile", q=[0.25, 0.75]),
+            pytest.param(
+                method("quantile", q=[0.25, 0.75]),
+                marks=pytest.mark.xfail(reason="nanquantile not implemented"),
+            ),
             method("reduce", func=np.sum, dim="x"),
             method("map", np.fabs),
         ),
@@ -5039,7 +5051,10 @@ def test_resample(self, dtype):
             method("assign_coords", v=("x", np.arange(10) * unit_registry.s)),
             method("first"),
             method("last"),
-            method("quantile", q=[0.25, 0.5, 0.75], dim="x"),
+            pytest.param(
+                method("quantile", q=[0.25, 0.5, 0.75], dim="x"),
+                marks=pytest.mark.xfail(reason="nanquantile not implemented"),
+            ),
         ),
         ids=repr,
     )