🧹 reduce dependencies (#572)

* Update requirements.txt

* make eofs not requirement

* Update climpred-dev.yml

* faster rm_poly

* rm xrft from requirements

* blacken notebooks

* rm tqdm from ci/requirements/docs_notebooks.yml

ci/requirements/climpred-dev.yml

@@ -18,6 +18,7 @@ dependencies:
   # Miscellaneous
   - lxml
   - tqdm
+  - cftime>=1.1.2 # Optimization changes to make cftime fast.
   # Numerics
   - numpy
   - pandas
@@ -49,13 +50,12 @@ dependencies:
   # Statistics
   - eofs
   - esmtools>=1.1.3
-  - xrft=0.2.0  # New version not compatible with xarray 0.16.1
   - xskillscore>=0.0.18
   # Visualization
   - matplotlib
   - nc-time-axis
   - pip
   - pip:
-      - cftime>=1.1.2 # Optimization changes to make cftime fast.
       - pytest-tldr
       - pytest-lazy-fixture
+      - nb_black  # notebook linting

ci/requirements/docs.yml

@@ -5,7 +5,6 @@ dependencies:
   - python=3.8
   - bottleneck
   - cartopy>=0.18.0
-  - esmtools>=1.1.3
   - importlib_metadata
   - jupyterlab
   - matplotlib

ci/requirements/docs_notebooks.yml

@@ -6,8 +6,6 @@ dependencies:
   - xesmf
   - esmpy
   - bottleneck
-  - cartopy>=0.18.0
-  - esmtools>=1.1.3
   - importlib_metadata
   - jupyterlab
   - matplotlib
@@ -22,9 +20,11 @@ dependencies:
   - sphinxcontrib-napoleon
   - sphinx_rtd_theme
   - toolz
-  - tqdm
   - xarray>=0.16.1
+  - xrft  # used for varweighted_mean_period
+  - esmtools>=1.1.3
   - pip
   - pip:
     # Install latest version of climpred.
     - -e ../..
+    - nb_black

ci/requirements/minimum-tests.yml

@@ -9,7 +9,6 @@ dependencies:
   - dask
   - eofs
   - esmpy=*=mpi*  # Ensures MPI works with version of esmpy.
-  - esmtools>=1.1.3
   - ipython
   - matplotlib
   - nc-time-axis

climpred/bootstrap.py

@@ -21,7 +21,12 @@
 from .metrics import ALL_METRICS, METRIC_ALIASES
 from .prediction import compute_hindcast, compute_perfect_model
 from .reference import compute_climatology, compute_persistence
-from .stats import dpp, varweighted_mean_period
+from .stats import dpp
+
+try:
+    from .stats import varweighted_mean_period
+except ImportError:
+    varweighted_mean_period = None
 from .utils import (
     _transpose_and_rechunk_to,
     assign_attrs,
@@ -1261,6 +1266,11 @@ def varweighted_mean_period_threshold(control, sig=95, iterations=500, time_dim=
         * climpred.bootstrap._bootstrap_func
         * climpred.stats.varweighted_mean_period
     """
+    if varweighted_mean_period is None:
+        raise ImportError(
+            "xrft is not installed; see"
+            "https://xrft.readthedocs.io/en/latest/installation.html"
+        )
     return _bootstrap_func(
         varweighted_mean_period,
         control,

climpred/relative_entropy.py

@@ -2,7 +2,11 @@
 
 import numpy as np
 import xarray as xr
-from eofs.xarray import Eof
+
+try:
+    from eofs.xarray import Eof
+except ImportError:
+    Eof = None
 
 
 def _relative_entropy_formula(sigma_b, sigma_x, mu_x, mu_b, neofs):
@@ -110,6 +114,11 @@ def compute_relative_entropy(
     Returns:
         rel_ent (xr.Dataset): relative entropy
     """
+    if Eof is None:
+        raise ImportError(
+            "eofs is not installed; see"
+            "https://ajdawson.github.io/eofs/latest/index.html"
+        )
     # Defaults
     if neofs is None:
         neofs = initialized.member.size

climpred/smoothing.py

@@ -61,7 +61,7 @@ def spatial_smoothing_xesmf(
     if xe is None:
         raise ImportError(
             "xesmf is not installed; see"
-            "https://xesmf.readthedocs.io/en/latest/installation.html"
+            "https://pangeo-xesmf.readthedocs.io/en/latest/installation.html"
         )
 
     def _regrid_it(da, d_lon, d_lat, **kwargs):

climpred/stats.py

@@ -2,22 +2,49 @@
 
 import numpy as np
 import xarray as xr
-from esmtools.stats import corr
-from xrft import power_spectrum
+from xskillscore import pearson_r
 
+try:
+    from xrft import power_spectrum
+except ImportError:
+    power_spectrum = None
 from .checks import is_xarray
 
 
+def rm_poly(ds, dim="time", deg=2, **kwargs):
+    """Remove degree polynomial along dimension dim from ds."""
+    coefficients = ds.polyfit(dim, deg=deg, **kwargs)
+    coord = ds[dim]
+    fits = []
+    if isinstance(ds, xr.Dataset):
+        for v in coefficients:
+            name = v.replace("_polyfit_coefficients", "")
+            fit = xr.polyval(coord, coefficients[v]).rename(name)
+            fits.append(fit)
+        fits = xr.merge(fits)
+    elif isinstance(ds, xr.DataArray):
+        name = ds.name
+        v = list(coefficients.data_vars)[0]
+        fits = xr.polyval(coord, coefficients[v]).rename(name)
+    ds_rm_poly = ds - fits
+    return ds_rm_poly
+
+
+def rm_trend(ds, dim="time", **kwargs):
+    """Remove degree polynomial along dimension dim from ds."""
+    return rm_poly(ds, dim=dim, deg=1, **kwargs)
+
+
 @is_xarray(0)
-def decorrelation_time(da, r=20, dim="time"):
+def decorrelation_time(da, iterations=20, dim="time"):
     """Calculate the decorrelaton time of a time series.
 
     .. math::
         \\tau_{d} = 1 + 2 * \\sum_{k=1}^{r}(\\alpha_{k})^{k}
 
     Args:
-        da (xarray object): Time series.
-        r (optional int): Number of iterations to run the above formula.
+        da (xarray object): input.
+        iterations (optional int): Number of iterations to run the above formula.
         dim (optional str): Time dimension for xarray object.
 
     Returns:
@@ -29,10 +56,20 @@ def decorrelation_time(da, r=20, dim="time"):
           p.373
 
     """
+
+    def _lag_corr(x, y, dim, lead):
+        """Helper function to shift the two time series and correlate."""
+        N = x[dim].size
+        normal = x.isel({dim: slice(0, N - lead)})
+        shifted = y.isel({dim: slice(0 + lead, N)})
+        # Align dimensions for xarray operation
+        shifted[dim] = normal[dim]
+        return pearson_r(normal, shifted, dim)
+
     one = xr.ones_like(da.isel({dim: 0}))
     one = one.where(da.isel({dim: 0}).notnull())
     return one + 2 * xr.concat(
-        [corr(da, da, dim=dim, lead=i) ** i for i in range(1, r)], "it"
+        [_lag_corr(da, da, dim=dim, lead=i) ** i for i in range(1, iterations)], "it"
     ).sum("it")
 
 
@@ -148,6 +185,11 @@ def varweighted_mean_period(da, dim="time", **kwargs):
     See also:
     https://xrft.readthedocs.io/en/latest/api.html#xrft.xrft.power_spectrum
     """
+    if power_spectrum is None:
+        raise ImportError(
+            "xrft is not installed; see"
+            "https://xrft.readthedocs.io/en/latest/installation.html"
+        )
     if isinstance(da, xr.Dataset):
         raise ValueError("require xr.DataArray, try xr.Dataset.map(func)")
     da = da.fillna(0.0)

climpred/tests/test_hindcast_prediction.py

@@ -5,18 +5,11 @@
 from climpred.bootstrap import bootstrap_hindcast
 from climpred.comparisons import HINDCAST_COMPARISONS
 from climpred.constants import CLIMPRED_DIMS
-from climpred.metrics import DETERMINISTIC_HINDCAST_METRICS
 from climpred.prediction import compute_hindcast
 from climpred.reference import compute_persistence
 
-# uacc is sqrt(MSSS), fails when MSSS negative
-DETERMINISTIC_HINDCAST_METRICS = DETERMINISTIC_HINDCAST_METRICS.copy()
-DETERMINISTIC_HINDCAST_METRICS.remove("uacc")
-
 ITERATIONS = 2
 
-category_edges = np.array([0, 0.5, 1])
-
 
 def test_compute_hindcast_lead0_lead1(
     hind_ds_initialized_1d, hind_ds_initialized_1d_lead0, reconstruction_ds_1d

climpred/tests/test_map.py

@@ -1,7 +1,7 @@
 import pytest
 import xarray as xr
-from esmtools.stats import rm_poly
 
+from climpred.stats import rm_poly
 from climpred.testing import assert_PredictionEnsemble
 
 xr.set_options(display_style="text")
@@ -16,7 +16,7 @@ def test_PredictionEnsemble_raises_error(hindcast_hist_obs_1d):
 def test_PredictionEnsemble_raises_warning(hindcast_hist_obs_1d):
     """Tests that PredictionEnsemble raises warning."""
     with pytest.warns(UserWarning):
-        hindcast_hist_obs_1d.map(rm_poly, dim="init", order=2)
+        hindcast_hist_obs_1d.map(rm_poly, dim="init", deg=2)
 
 
 def test_PredictionEnsemble_xr_calls(hindcast_hist_obs_1d):
@@ -31,12 +31,12 @@ def test_PredictionEnsemble_xr_calls(hindcast_hist_obs_1d):
 def test_PredictionEnsemble_map_dim_or(hindcast_hist_obs_1d):
     """Tests that PredictionEnsemble allows dim0_or_dim1 as kwargs without UserWarning."""
     with pytest.warns(None):  # no warnings
-        he_or = hindcast_hist_obs_1d.map(rm_poly, dim="init_or_time", order=2)
+        he_or = hindcast_hist_obs_1d.map(rm_poly, dim="init_or_time", deg=2)
     assert he_or != hindcast_hist_obs_1d
 
     with pytest.warns(UserWarning) as record:  # triggers warnings
-        he_chained = hindcast_hist_obs_1d.map(rm_poly, dim="init", order=2).map(
-            rm_poly, dim="time", order=2
+        he_chained = hindcast_hist_obs_1d.map(rm_poly, dim="init", deg=2).map(
+            rm_poly, dim="time", deg=2
         )
     assert he_chained != hindcast_hist_obs_1d
 
@@ -52,4 +52,4 @@ def test_PredictionEnsemble_map_dim_or_fails_if_both_dims_in_dataset(
 ):
     """Tests that PredictionEnsemble with dim0_or_dim1 as kwargs fails if both dims in any dataset."""
     with pytest.raises(ValueError, match="cannot be both in"):
-        hindcast_hist_obs_1d.map(rm_poly, dim="init_or_lead", order=2)
+        hindcast_hist_obs_1d.map(rm_poly, dim="init_or_lead", deg=2)

climpred/tests/test_relative_entropy.py

@@ -1,10 +1,20 @@
+import pytest
+
 from climpred.graphics import plot_relative_entropy
 from climpred.relative_entropy import (
     bootstrap_relative_entropy,
     compute_relative_entropy,
 )
 
+try:
+    from eofs.xarray import Eof
+
+    Eof_loaded = True
+except ImportError:
+    Eof_loaded = False
+
 
+@pytest.mark.skipif(not Eof_loaded, reason="eofs not installed")
 def test_compute_relative_entropy(PM_da_initialized_3d, PM_da_control_3d):
     """
     Checks that there are no NaNs.
@@ -17,6 +27,7 @@ def test_compute_relative_entropy(PM_da_initialized_3d, PM_da_control_3d):
         assert not actual_any_nan[var]
 
 
+@pytest.mark.skipif(not Eof_loaded, reason="eofs not installed")
 def test_bootstrap_relative_entropy(PM_da_initialized_3d, PM_da_control_3d):
     """
     Checks that there are no NaNs.
@@ -33,6 +44,7 @@ def test_bootstrap_relative_entropy(PM_da_initialized_3d, PM_da_control_3d):
         assert not actual_any_nan[var]
 
 
+@pytest.mark.skipif(not Eof_loaded, reason="eofs not installed")
 def test_plot_relative_entropy(PM_da_initialized_3d, PM_da_control_3d):
     res = compute_relative_entropy(
         PM_da_initialized_3d, PM_da_control_3d, nmember_control=5, neofs=2

climpred/tests/test_stats.py

@@ -3,8 +3,16 @@
 import xarray as xr
 from xarray.testing import assert_allclose
 
-from climpred.bootstrap import dpp_threshold, varweighted_mean_period_threshold
-from climpred.stats import decorrelation_time, dpp, varweighted_mean_period
+from climpred.bootstrap import dpp_threshold
+from climpred.stats import decorrelation_time, dpp
+
+try:
+    from climpred.bootstrap import varweighted_mean_period_threshold
+    from climpred.stats import varweighted_mean_period
+
+    xrft_loaded = True
+except ImportError:
+    xrft_loaded = False
 
 ITERATIONS = 2
 
@@ -16,6 +24,7 @@ def test_dpp(PM_da_control_3d, chunk):
     assert res.mean() > 0
 
 
+@pytest.mark.skipif(not xrft_loaded, reason="xrft not installed")
 @pytest.mark.parametrize("func", (varweighted_mean_period, decorrelation_time))
 def test_potential_predictability_likely(PM_da_control_3d, func):
     """Check for positive diagnostic potential predictability in NA SST."""
@@ -32,6 +41,7 @@ def test_bootstrap_dpp_sig50_similar_dpp(PM_da_control_3d):
     xr.testing.assert_allclose(actual, expected, atol=0.5, rtol=0.5)
 
 
+@pytest.mark.skipif(not xrft_loaded, reason="xrft not installed")
 def test_bootstrap_vwmp_sig50_similar_vwmp(PM_da_control_3d):
     sig = 50
     actual = varweighted_mean_period_threshold(
@@ -48,6 +58,7 @@ def test_bootstrap_func_multiple_sig_levels(PM_da_control_3d):
     assert (actual.isel(quantile=0).values <= actual.isel(quantile=1)).all()
 
 
+@pytest.mark.skipif(not xrft_loaded, reason="xrft not installed")
 @pytest.mark.parametrize("step", [1, 2, -1])
 @pytest.mark.parametrize(
     "func",