Merge pull request #76 from theislab/dev

Merge dev into master v0.5.0

Author: davidsebfischer

diffxpy/__init__.py

@@ -1,4 +1,6 @@
-
 from ._version import get_versions
+
 __version__ = get_versions()['version']
 del get_versions
+
+from .log_cfg import logger, unconfigure_logging, enable_logging

diffxpy/api/__init__.py

@@ -1,5 +1,7 @@
+from .. import __version__
+from ..log_cfg import logger, unconfigure_logging, enable_logging
+
 from . import test
 from . import enrich
 from . import stats
-
-from .. import __version__
+from . import utils

diffxpy/api/enrich.py

@@ -1,2 +1,2 @@
-from ..enrichment.enrich import RefSets
-from ..enrichment.enrich import test
+from diffxpy.enrichment.enrich import RefSets
+from diffxpy.enrichment.enrich import test

diffxpy/api/stats.py

@@ -1,8 +1,10 @@
-from ..stats.stats import hypergeom_test
-from ..stats.stats import likelihood_ratio_test
-from ..stats.stats import wald_test
-from ..stats.stats import wald_test_chisq
-from ..stats.stats import two_coef_z_test
-from ..stats.stats import wilcoxon_test
-from ..stats.stats import t_test_moments
-from ..stats.stats import t_test_raw
+from diffxpy.stats.stats import hypergeom_test
+from diffxpy.stats.stats import likelihood_ratio_test
+from diffxpy.stats.stats import wald_test
+from diffxpy.stats.stats import wald_test_chisq
+from diffxpy.stats.stats import two_coef_z_test
+from diffxpy.stats.stats import wilcoxon_test
+from diffxpy.stats.stats import t_test_moments
+from diffxpy.stats.stats import t_test_raw
+
+from ..testing.correction import correct

diffxpy/api/test.py

@@ -1,11 +1,11 @@
-# from ..testing.base import stats
-from ..testing.base import two_sample
-from ..testing.base import lrt
-from ..testing.base import wald
-from ..testing.base import t_test
-from ..testing.base import wilcoxon
-from ..testing.base import partition
-from ..testing.base import pairwise
-from ..testing.base import versus_rest 
-from ..testing.base import design_matrix 
-from ..testing.base import coef_names 
+from diffxpy.testing.base import two_sample
+from diffxpy.testing.base import lrt
+from diffxpy.testing.base import wald
+from diffxpy.testing.base import t_test
+from diffxpy.testing.base import wilcoxon
+from diffxpy.testing.base import partition
+from diffxpy.testing.base import pairwise
+from diffxpy.testing.base import versus_rest
+from diffxpy.testing.base import continuous_1d
+from diffxpy.testing.base import design_matrix
+from diffxpy.testing.base import coef_names

diffxpy/api/utils.py

@@ -0,0 +1 @@
+import batchglm.data as data_utils

diffxpy/log_cfg.py

@@ -0,0 +1,31 @@
+import sys
+
+import logging
+
+logger = logging.getLogger('.'.join(__name__.split('.')[:-1]))
+
+_is_interactive = bool(getattr(sys, 'ps1', sys.flags.interactive))
+_hander = None
+
+
+def unconfigure_logging():
+    if _hander is not None:
+        logger.removeHandler(_hander)
+
+    logger.setLevel(logging.NOTSET)
+
+
+def enable_logging(verbosity=logging.ERROR, stream=sys.stderr, format=logging.BASIC_FORMAT):
+    unconfigure_logging()
+
+    logger.setLevel(verbosity)
+    _handler = logging.StreamHandler(stream)
+    _handler.setFormatter(logging.Formatter(format, None))
+    logger.addHandler(_handler)
+
+
+# If we are in an interactive environment (like Jupyter), set loglevel to INFO and pipe the output to stdout.
+if _is_interactive:
+    enable_logging(logging.INFO, sys.stdout)
+else:
+    enable_logging(logging.WARNING, sys.stderr)

diffxpy/pkg_constants.py

@@ -0,0 +1,7 @@
+BATCHGLM_OPTIM_GD = True
+BATCHGLM_OPTIM_ADAM = True
+BATCHGLM_OPTIM_ADAGRAD = False
+BATCHGLM_OPTIM_RMSPROP = False
+BATCHGLM_OPTIM_NEWTON = True
+BATCHGLM_OPTIM_IRLS = True
+BATCHGLM_TERMINATION_TYPE = "by_feature"

diffxpy/stats/stats.py

@@ -1,4 +1,5 @@
 import numpy as np
+import numpy.linalg
 import scipy.stats
 from typing import Union
 
@@ -218,7 +219,7 @@ def wald_test(
 
 def wald_test_chisq(
         theta_mle: np.ndarray,
-        theta_invcovar: np.ndarray,
+        theta_covar: np.ndarray,
         theta0: Union[int, np.ndarray] = 0
 ):
     """
@@ -236,8 +237,8 @@ def wald_test_chisq(
 
     :param theta_mle: np.array (par x genes)
         Maximum likelihood estimator of given parameter by gene.
-    :param theta_invcovar:  np.array (genes x par x par)
-        Inverse of the covariance matrix of the parameters in theta_mle by gene.
+    :param theta_covar:  np.array (genes x par x par)
+        Covariance matrix of the parameters in theta_mle by gene.
         This is the negative hessian or the inverse of the 
         observed fisher information matrix.
     :param theta0: float
@@ -246,13 +247,13 @@ def wald_test_chisq(
     if np.size(theta0) == 1:
         theta0 = np.broadcast_to(theta0, theta_mle.shape)
 
-    if theta_mle.shape[0] != theta_invcovar.shape[1]:
+    if theta_mle.shape[0] != theta_covar.shape[1]:
         raise ValueError(
-            'stats.wald_test(): theta_mle and theta_invcovar have to contain the same number of parameters')
-    if theta_mle.shape[1] != theta_invcovar.shape[0]:
-        raise ValueError('stats.wald_test(): theta_mle and theta_invcovar have to contain the same number of genes')
-    if theta_invcovar.shape[1] != theta_invcovar.shape[2]:
-        raise ValueError('stats.wald_test(): the first two dimensions of theta_invcovar have to be of the same size')
+            'stats.wald_test(): theta_mle and theta_covar have to contain the same number of parameters')
+    if theta_mle.shape[1] != theta_covar.shape[0]:
+        raise ValueError('stats.wald_test(): theta_mle and theta_covar have to contain the same number of genes')
+    if theta_covar.shape[1] != theta_covar.shape[2]:
+        raise ValueError('stats.wald_test(): the first two dimensions of theta_covar have to be of the same size')
     if theta0.shape[0] > 1:
         if theta_mle.shape[0] != theta0.shape[0]:
             raise ValueError('stats.wald_test(): theta_mle and theta0 have to contain the same number of entries')
@@ -264,7 +265,7 @@ def wald_test_chisq(
         np.matmul(
             np.matmul(
                 theta_diff[:, [i]].T,
-                theta_invcovar[i, :, :]
+                numpy.linalg.inv(theta_covar[i, :, :])
             ),
             theta_diff[:, [i]]
         )