fixed new single tests by noise model

Author: davidsebfischer

diffxpy/testing/tests.py

@@ -19,8 +19,6 @@
     DifferentialExpressionTestWaldCont, DifferentialExpressionTestLRTCont
 from .utils import parse_gene_names, parse_data, parse_sample_description, parse_size_factors, parse_grouping
 
-logger = logging.getLogger("diffxpy")
-
 # Use this to suppress matrix subclass PendingDepreceationWarnings from numpy:
 np.warnings.filterwarnings("ignore")
 
@@ -145,8 +143,8 @@ def _fit(
         else:
             raise ValueError('base.test(): `noise_model="%s"` not recognized.' % noise_model)
 
-        logger.info("Fitting model...")
-        logger.debug(" * Assembling input data...")
+        logging.getLogger("diffxpy").info("Fitting model...")
+        logging.getLogger("diffxpy").debug(" * Assembling input data...")
         input_data = InputData.new(
             data=data,
             design_loc=design_loc,
@@ -157,7 +155,7 @@ def _fit(
             feature_names=gene_names,
         )
 
-        logger.debug(" * Set up Estimator...")
+        logging.getLogger("diffxpy").debug(" * Set up Estimator...")
         constructor_args = {}
         if batch_size is not None:
             constructor_args["batch_size"] = batch_size
@@ -176,10 +174,10 @@ def _fit(
             **constructor_args
         )
 
-        logger.debug(" * Initializing Estimator...")
+        logging.getLogger("diffxpy").debug(" * Initializing Estimator...")
         estim.initialize()
 
-        logger.debug(" * Run estimation...")
+        logging.getLogger("diffxpy").debug(" * Run estimation...")
         # training:
         if callable(training_strategy):
             # call training_strategy if it is a function
@@ -188,11 +186,11 @@ def _fit(
             estim.train_sequence(training_strategy=training_strategy)
 
         if close_session:
-            logger.debug(" * Finalize estimation...")
+            logging.getLogger("diffxpy").debug(" * Finalize estimation...")
             model = estim.finalize()
         else:
             model = estim
-        logger.debug(" * Model fitting done.")
+        logging.getLogger("diffxpy").debug(" * Model fitting done.")
 
     return model
 

diffxpy/unit_test/test_single_de.py

@@ -4,18 +4,30 @@
 import pandas as pd
 import scipy.stats as stats
 
-from batchglm.api.models.glm_nb import Simulator
 import diffxpy.api as de
 
 
 class _TestSingleDE:
 
-    def _prepare_data(self, n_cells: int = 2000, n_genes: int = 100):
+    def _prepare_data(
+            self,
+            n_cells: int,
+            n_genes: int,
+            noise_model: str
+    ):
         """
 
         :param n_cells: Number of cells to simulate (number of observations per test).
         :param n_genes: Number of genes to simulate (number of tests).
+        :param noise_model: Noise model to use for data fitting.
         """
+        if noise_model == "nb":
+            from batchglm.api.models.glm_nb import Simulator
+        elif noise_model == "norm":
+            from batchglm.api.models.glm_norm import Simulator
+        else:
+            raise ValueError("noise model %s not recognized" % noise_model)
+
         num_non_de = n_genes // 2
         sim = Simulator(num_observations=n_cells, num_features=n_genes)
         sim.generate_sample_description(num_batches=0, num_conditions=2)
@@ -50,7 +62,12 @@ def _eval(self, sim, test):
 
         return sim
 
-    def _test_rank_de(self, n_cells: int = 2000, n_genes: int = 100):
+    def _test_rank_de(
+            self,
+            n_cells: int,
+            n_genes: int,
+            noise_model: str
+    ):
         """
         :param n_cells: Number of cells to simulate (number of observations per test).
         :param n_genes: Number of genes to simulate (number of tests).
@@ -59,7 +76,11 @@ def _test_rank_de(self, n_cells: int = 2000, n_genes: int = 100):
         logging.getLogger("batchglm").setLevel(logging.WARNING)
         logging.getLogger("diffxpy").setLevel(logging.WARNING)
 
-        sim = self._prepare_data(n_cells=n_cells, n_genes=n_genes)
+        sim = self._prepare_data(
+            n_cells=n_cells,
+            n_genes=n_genes,
+            noise_model=noise_model
+        )
 
         test = de.test.rank_test(
             data=sim.X,
@@ -72,7 +93,11 @@ def _test_rank_de(self, n_cells: int = 2000, n_genes: int = 100):
 
         return True
 
-    def _test_t_test_de(self, n_cells: int = 2000, n_genes: int = 100):
+    def _test_t_test_de(
+            self,
+            n_cells: int,
+            n_genes: int
+    ):
         """
         :param n_cells: Number of cells to simulate (number of observations per test).
         :param n_genes: Number of genes to simulate (number of tests).
@@ -81,7 +106,11 @@ def _test_t_test_de(self, n_cells: int = 2000, n_genes: int = 100):
         logging.getLogger("batchglm").setLevel(logging.WARNING)
         logging.getLogger("diffxpy").setLevel(logging.WARNING)
 
-        sim = self._prepare_data(n_cells=n_cells, n_genes=n_genes)
+        sim = self._prepare_data(
+            n_cells=n_cells,
+            n_genes=n_genes,
+            noise_model="norm"
+        )
 
         test = de.test.t_test(
             data=sim.X,
@@ -109,7 +138,11 @@ def _test_wald_de(
         logging.getLogger("batchglm").setLevel(logging.WARNING)
         logging.getLogger("diffxpy").setLevel(logging.WARNING)
 
-        sim = self._prepare_data(n_cells=n_cells, n_genes=n_genes)
+        sim = self._prepare_data(
+            n_cells=n_cells,
+            n_genes=n_genes,
+            noise_model=noise_model
+        )
 
         test = de.test.wald(
             data=sim.X,
@@ -140,7 +173,11 @@ def _test_lrt_de(
         logging.getLogger("batchglm").setLevel(logging.WARNING)
         logging.getLogger("diffxpy").setLevel(logging.WARNING)
 
-        sim = self._prepare_data(n_cells=n_cells, n_genes=n_genes)
+        sim = self._prepare_data(
+            n_cells=n_cells,
+            n_genes=n_genes,
+            noise_model=noise_model
+        )
 
         test = de.test.lrt(
             data=sim.X,
@@ -264,5 +301,6 @@ def test_lrt_de_norm(
             noise_model="norm"
         )
 
+
 if __name__ == '__main__':
     unittest.main()