Merge pull request #83 from theislab/dev

Dev v0.6.5

Author: davidsebfischer

batchglm/api/data.py

@@ -1,3 +1,4 @@
 from batchglm.data import design_matrix
-from batchglm.data import constraint_matrix_from_dict, constraint_matrix_from_string, string_constraints_from_dict
+from batchglm.data import constraint_matrix_from_dict, constraint_matrix_from_string, string_constraints_from_dict, \
+    constraint_system_from_star
 from batchglm.data import view_coef_names, preview_coef_names

batchglm/api/typing.py

@@ -1,2 +1,2 @@
 from batchglm.models.base.estimator import EstimatorBaseTyping
-from batchglm.models.base.input import InputDataBaseTyping
+from batchglm.models.base.input import InputDataBase

batchglm/data.py

@@ -1,11 +1,8 @@
-import dask
-import dask.array
 import logging
 import patsy
 import pandas as pd
 import numpy as np
-import scipy.sparse
-from typing import Union, Dict, Tuple, List
+from typing import Union, Tuple, List
 
 try:
     import anndata
@@ -23,7 +20,7 @@ def design_matrix(
         formula: Union[str, None] = None,
         as_categorical: Union[bool, list] = True,
         dmat: Union[pd.DataFrame, None] = None,
-        return_type: str = "xarray",
+        return_type: str = "patsy",
 ) -> Union[patsy.design_info.DesignMatrix, pd.DataFrame]:
     """
     Create a design matrix from some sample description.
@@ -142,7 +139,7 @@ def constraint_system_from_star(
         formula: Union[None, str] = None,
         as_categorical: Union[bool, list] = True,
         constraints: Union[None, List[str], Tuple[str], dict, np.ndarray] = None,
-        return_type: str = "xarray",
+        return_type: str = "patsy",
 ) -> Tuple:
     """
     Wrap different constraint matrix building formats with building of design matrix.

batchglm/models/base/__init__.py

@@ -1,4 +1,4 @@
-from .input import _InputDataBase, InputDataBaseTyping
+from .input import InputDataBase, InputDataBase
 from .estimator import _EstimatorBase, EstimatorBaseTyping
 from .model import _ModelBase
 from .simulator import _SimulatorBase

batchglm/models/base/estimator.py

@@ -9,7 +9,7 @@
 except ImportError:
     anndata = None
 
-from .input import _InputDataBase
+from .input import InputDataBase
 from .model import _ModelBase
 
 logger = logging.getLogger(__name__)
@@ -29,7 +29,7 @@ class TrainingStrategy(Enum):
     def __init__(
             self,
             model: _ModelBase,
-            input_data: _InputDataBase
+            input_data: InputDataBase
     ):
         self.model = model
         self.input_data = input_data

batchglm/models/base/input.py

@@ -11,7 +11,7 @@
 logger = logging.getLogger(__name__)
 
 
-class _InputDataBase:
+class InputDataBase:
     """
     Base class for all input data types.
     """
@@ -45,7 +45,7 @@ def __init__(
             self.x = data
         elif isinstance(data, anndata.AnnData) or isinstance(data, anndata.Raw):
             self.x = data.X
-        elif isinstance(data, _InputDataBase):
+        elif isinstance(data, InputDataBase):
             self.x = data.x
         else:
             raise ValueError("type of data %s not recognized" % type(data))
@@ -90,8 +90,3 @@ def fetch_x_sparse(self, idx):
             data_idx = np.squeeze(data_idx, axis=0)
 
         return data_idx, data_val, data_shape
-
-class InputDataBaseTyping:
-    """
-    Input data base class  used for typing in other packages.
-    """

batchglm/models/base/simulator.py

@@ -8,7 +8,7 @@
 except ImportError:
     anndata = None
 
-from .input import _InputDataBase
+from .input import InputDataBase
 from .model import _ModelBase
 
 logger = logging.getLogger(__name__)
@@ -26,7 +26,7 @@ class _SimulatorBase(metaclass=abc.ABCMeta):
     nobs: int
     nfeatures: int
 
-    input_data: _InputDataBase
+    input_data: InputDataBase
     model: _ModelBase
 
     def __init__(

batchglm/models/base_glm/external.py

@@ -1,5 +1,5 @@
 from batchglm.models.base import _EstimatorBase
-from batchglm.models.base import _InputDataBase
+from batchglm.models.base import InputDataBase
 from batchglm.models.base import _ModelBase
 from batchglm.models.base import _SimulatorBase
 

batchglm/models/base_glm/input.py

@@ -10,10 +10,10 @@
 from typing import Union
 
 from .utils import parse_constraints, parse_design
-from .external import _InputDataBase
+from .external import InputDataBase
 
 
-class InputDataGLM(_InputDataBase):
+class InputDataGLM(InputDataBase):
     """
     Input data for Generalized Linear Models (GLMs).
     """
@@ -76,7 +76,7 @@ def __init__(
             If this option is set, all provided data will be casted to this data type.
         :return: InputData object
         """
-        _InputDataBase.__init__(
+        InputDataBase.__init__(
             self=self,
             data=data,
             observation_names=observation_names,

batchglm/train/tf/glm_beta/estimator.py

@@ -20,7 +20,7 @@ class Estimator(TFEstimatorGLM, ProcessModel):
     def __init__(
             self,
             input_data: InputDataGLM,
-            batch_size: int = 500,
+            batch_size: int = 512,
             graph: tf.Graph = None,
             init_model: Model = None,
             init_a: Union[np.ndarray, str] = "AUTO",