flake8 passing

deicode/q2/plugin_setup.py

@@ -12,7 +12,7 @@
 from deicode.rpca import rpca, auto_rpca
 from deicode._rpca_defaults import (DESC_RANK, DESC_MSC, DESC_MFC,
                                     DESC_ITERATIONS, DESC_MFF)
-from qiime2.plugin import (Properties, Int, Float, Choices, Str)
+from qiime2.plugin import (Properties, Int, Float)
 from q2_types.feature_table import FeatureTable, Frequency
 from q2_types.distance_matrix import DistanceMatrix
 from q2_types.ordination import PCoAResults

deicode/q2/tests/test_method.py

@@ -49,6 +49,18 @@ def test_rpca(self):
         self.assertFalse(np.isnan(ord_test.features).any(axis=None))
         self.assertFalse(np.isnan(ord_test.samples).any(axis=None))
 
+    def test_auto_rpca(self):
+        """Tests the basic validity of the actual auto_rpca()."""
+        ord_test, dist_test = auto_rpca(table=self.test_table)
+        # Validate types of the RPCA outputs
+        self.assertIsInstance(ord_test, OrdinationResults)
+        self.assertIsInstance(dist_test, DistanceMatrix)
+        # Ensure that no NaNs are in the OrdinationResults
+        # NOTE that we have to use the DataFrame .any() functions instead of
+        # python's built-in any() functions -- see #29 for details on this
+        self.assertFalse(np.isnan(ord_test.features).any(axis=None))
+        self.assertFalse(np.isnan(ord_test.samples).any(axis=None))
+
 
 class Test_qiime2_rpca(unittest.TestCase):
 

deicode/rpca.py

@@ -54,7 +54,7 @@ def frequency_filter(val, id_, md):
     # get new n-comp when applicable
     n_components = opt.s.shape[0]
     # get PC column labels for the skbio OrdinationResults
-    rename_cols = ['PC' + str(i+1) for i in range(n_components)]
+    rename_cols = ['PC' + str(i + 1) for i in range(n_components)]
     # get completed matrix for centering
     X = opt.sample_weights @ opt.s @ opt.feature_weights.T
     # center again around zero after completion
@@ -106,20 +106,21 @@ def frequency_filter(val, id_, md):
 
     return ord_res, dist_res
 
+
 def auto_rpca(table: biom.Table,
               min_sample_count: int = DEFAULT_MSC,
               min_feature_count: int = DEFAULT_MFC,
               min_feature_frequency: float = DEFAULT_MFF,
               max_iterations: int = DEFAULT_ITERATIONS) -> (
-              skbio.OrdinationResults,
-              skbio.DistanceMatrix):
+        skbio.OrdinationResults,
+        skbio.DistanceMatrix):
     """Runs RPCA but with auto estimation of the
        rank peramater.
     """
     ord_res, dist_res = rpca(table,
-                             n_components = 'auto',
-                             min_sample_count = min_sample_count,
-                             min_feature_count = min_feature_count,
-                             min_feature_frequency = min_feature_frequency,
-                             max_iterations = max_iterations)
+                             n_components='auto',
+                             min_sample_count=min_sample_count,
+                             min_feature_count=min_feature_count,
+                             min_feature_frequency=min_feature_frequency,
+                             max_iterations=max_iterations)
     return ord_res, dist_res

deicode/scripts/_standalone_rpca.py

@@ -1,6 +1,5 @@
 import os
 import click
-from typing import Union
 from biom import load_table
 from deicode.rpca import rpca as _rpca
 from deicode.rpca import auto_rpca as _auto_rpca
@@ -14,38 +13,34 @@
 def deicode():
     pass
 
+
 @deicode.command('rpca')
 @click.option('--in-biom',
               help='Input table in biom format.',
               required=True)
 @click.option('--output-dir',
               help='Location of output files.',
               required=True)
-@click.option(
-    '--n_components',
-    default=DEFAULT_RANK,
-    show_default=True,
-    help=DESC_RANK)
-@click.option(
-    '--min-sample-count',
-    default=DEFAULT_MSC,
-    show_default=True,
-    help=DESC_MSC)
-@click.option(
-    '--min-feature-count',
-    default=DEFAULT_MFC,
-    show_default=True,
-    help=DESC_MFC)
-@click.option(
-    '--min-feature-frequency',
-    default=DEFAULT_MFF,
-    show_default=True,
-    help=DESC_MFF)
-@click.option(
-    '--max_iterations',
-    default=DEFAULT_ITERATIONS,
-    show_default=True,
-    help=DESC_ITERATIONS)
+@click.option('--n_components',
+              default=DEFAULT_RANK,
+              show_default=True,
+              help=DESC_RANK)
+@click.option('--min-sample-count',
+              default=DEFAULT_MSC,
+              show_default=True,
+              help=DESC_MSC)
+@click.option('--min-feature-count',
+              default=DEFAULT_MFC,
+              show_default=True,
+              help=DESC_MFC)
+@click.option('--min-feature-frequency',
+              default=DEFAULT_MFF,
+              show_default=True,
+              help=DESC_MFF)
+@click.option('--max_iterations',
+              default=DEFAULT_ITERATIONS,
+              show_default=True,
+              help=DESC_ITERATIONS)
 def rpca(in_biom: str,
          output_dir: str,
          n_components: int,
@@ -59,11 +54,11 @@ def rpca(in_biom: str,
     table = load_table(in_biom)
     # run the RPCA wrapper
     ord_res, dist_res = _rpca(table,
-                             n_components,
-                             min_sample_count,
-                             min_feature_count,
-                             min_feature_frequency,
-                             max_iterations)
+                              n_components,
+                              min_sample_count,
+                              min_feature_count,
+                              min_feature_frequency,
+                              max_iterations)
 
     # If it doesn't already exist, create the output directory.
     # Note that there is technically a race condition here: it's ostensibly
@@ -81,6 +76,7 @@ def rpca(in_biom: str,
     ord_res.write(os.path.join(output_dir, 'ordination.txt'))
     dist_res.write(os.path.join(output_dir, 'distance-matrix.tsv'))
 
+
 @deicode.command('auto-rpca')
 @click.option('--in-biom',
               help='Input table in biom format.',
@@ -121,10 +117,10 @@ def auto_rpca(in_biom: str,
     table = load_table(in_biom)
     # run the RPCA wrapper
     ord_res, dist_res = _auto_rpca(table,
-                                  min_sample_count,
-                                  min_feature_count,
-                                  min_feature_frequency,
-                                  max_iterations)
+                                   min_sample_count,
+                                   min_feature_count,
+                                   min_feature_frequency,
+                                   max_iterations)
 
     # If it doesn't already exist, create the output directory.
     # Note that there is technically a race condition here: it's ostensibly

deicode/scripts/tests/test_standalone_rpca.py

@@ -22,8 +22,8 @@ def test_standalone_rpca_rank_est(self):
         out_ = os_path_sep.join(in_.split(os_path_sep)[:-1])
         runner = CliRunner()
         result = runner.invoke(sdc.commands['auto-rpca'],
-                                ['--in-biom', in_,
-                                 '--output-dir', out_])
+                               ['--in-biom', in_,
+                                '--output-dir', out_])
         # Read the results
         dist_res = pd.read_csv(get_data_path('distance-matrix.tsv'), sep='\t',
                                index_col=0)

deicode/testing.py

@@ -1,7 +1,3 @@
-import numpy as np
-from scipy.stats import norm
-from skbio.stats.composition import closure
-from numpy.random import poisson, lognormal, normal
 from pandas.testing import assert_series_equal
 
 

deicode/tests/test_optspace.py

@@ -8,7 +8,6 @@
     OptSpace,
     svd_sort,
     rank_estimate)
-from deicode.preprocessing import rclr
 import numpy as np
 from numpy.linalg import norm
 import unittest