refactor: separate mcar, mar, mnar missing patterns;

pycorruptor/__init__.py

@@ -1,5 +1,5 @@
 """
-PyCorruptor package
+PyCorruptor package.
 """
 
 # Created by Wenjie Du <[email protected]>
@@ -24,12 +24,11 @@
 __version__ = "0.0.4"
 
 try:
-    from pycorruptor.corrupt import (
+    from pycorruptor.mcar import mcar
+    from pycorruptor.utils import (
         cal_missing_rate,
         masked_fill,
-        mcar,
     )
-
 except Exception as e:
     print(e)
 

pycorruptor/mar.py

@@ -0,0 +1,43 @@
+"""
+Corrupt data by adding missing values to it with MAR (missing at random) pattern.
+"""
+
+# Created by Wenjie Du <[email protected]>
+# License: GLP-v3
+
+
+def mar(X, rate, nan=0):
+    """Create random missing values (MAR case).
+
+    Parameters
+    ----------
+    X : array,
+        Data vector. If X has any missing values, they should be numpy.nan.
+
+    rate : float, in (0,1),
+        Artificially missing rate, rate of the observed values which will be artificially masked as missing.
+
+        Note that,
+        `rate` = (number of artificially missing values) / np.sum(~np.isnan(self.data)),
+        not (number of artificially missing values) / np.product(self.data.shape),
+        considering that the given data may already contain missing values,
+        the latter way may be confusing because if the original missing rate >= `rate`,
+        the function will do nothing, i.e. it won't play the role it has to be.
+
+    nan : int/float, optional, default=0
+        Value used to fill NaN values.
+
+    Returns
+    -------
+
+    """
+    # TODO: Create missing values in MAR case
+    raise NotImplementedError("MAR case has not been implemented yet.")
+
+
+def _mar_numpy(X, rate, nan=0):
+    pass
+
+
+def _mar_torch(X, rate, nan=0):
+    pass

pycorruptor/corrupt.py → pycorruptor/mcar.py

@@ -1,5 +1,5 @@
 """
-Corrupt data by adding missing values to it with optional missing patterns (MCAR,MAR,MNAR).
+Corrupt data by adding missing values to it with MCAR (missing completely at random) pattern.
 """
 
 # Created by Wenjie Du <[email protected]>
@@ -13,89 +13,6 @@
     pass
 
 
-def cal_missing_rate(X):
-    """Calculate the originally missing rate of the raw data.
-
-    Parameters
-    ----------
-    X : array-like,
-        Data array that may contain missing values.
-
-    Returns
-    -------
-    originally_missing_rate, float,
-        The originally missing rate of the raw data.
-    """
-    if isinstance(X, list):
-        X = np.asarray(X)
-
-    if isinstance(X, np.ndarray):
-        originally_missing_rate = np.sum(np.isnan(X)) / np.product(X.shape)
-    elif isinstance(X, torch.Tensor):
-        originally_missing_rate = torch.sum(torch.isnan(X)) / np.product(X.shape)
-        originally_missing_rate = originally_missing_rate.item()
-    else:
-        raise TypeError(
-            "X must be type of list/numpy.ndarray/torch.Tensor, " f"but got {type(X)}"
-        )
-
-    return originally_missing_rate
-
-
-def masked_fill(X, mask, val):
-    """Like torch.Tensor.masked_fill(), fill elements in given `X` with `val` where `mask` is True.
-
-    Parameters
-    ----------
-    X : array-like,
-        The data vector.
-
-    mask : array-like,
-        The boolean mask.
-
-    val : float
-        The value to fill in with.
-
-    Returns
-    -------
-    array,
-        mask
-    """
-    assert X.shape == mask.shape, (
-        "Shapes of X and mask must match, "
-        f"but X.shape={X.shape}, mask.shape={mask.shape}"
-    )
-    assert isinstance(X, type(mask)), (
-        "Data types of X and mask must match, " f"but got {type(X)} and {type(mask)}"
-    )
-
-    if isinstance(X, list):
-        X = np.asarray(X)
-        mask = np.asarray(mask)
-
-    if isinstance(X, np.ndarray):
-        mask = mask.astype(bool)
-        X[mask] = val
-    elif isinstance(X, torch.Tensor):
-        mask = mask.type(torch.bool)
-        X[mask] = val
-    else:
-        raise TypeError(
-            "X must be type of list/numpy.ndarray/torch.Tensor, " f"but got {type(X)}"
-        )
-
-    return X
-
-
-def little_mcar_test(X):
-    """Little's MCAR Test.
-
-    Refer to :cite:`little1988TestMCAR`
-    """
-    # TODO: Little's MCAR test
-    raise NotImplementedError("MCAR test has not been implemented yet.")
-
-
 def mcar(X, rate, nan=0):
     """Create completely random missing values (MCAR case).
 
@@ -195,59 +112,10 @@ def _mcar_torch(X, rate, nan=0):
     return X_intact, X, missing_mask, indicating_mask
 
 
-def mar(X, rate, nan=0):
-    """Create random missing values (MAR case).
-
-    Parameters
-    ----------
-    X : array,
-        Data vector. If X has any missing values, they should be numpy.nan.
-
-    rate : float, in (0,1),
-        Artificially missing rate, rate of the observed values which will be artificially masked as missing.
-
-        Note that,
-        `rate` = (number of artificially missing values) / np.sum(~np.isnan(self.data)),
-        not (number of artificially missing values) / np.product(self.data.shape),
-        considering that the given data may already contain missing values,
-        the latter way may be confusing because if the original missing rate >= `rate`,
-        the function will do nothing, i.e. it won't play the role it has to be.
-
-    nan : int/float, optional, default=0
-        Value used to fill NaN values.
-
-    Returns
-    -------
-
-    """
-    # TODO: Create missing values in MAR case
-    raise NotImplementedError("MAR case has not been implemented yet.")
-
-
-def mnar(X, rate, nan=0):
-    """Create not-random missing values (MNAR case).
-
-    Parameters
-    ----------
-    X : array,
-        Data vector. If X has any missing values, they should be numpy.nan.
-
-    rate : float, in (0,1),
-        Artificially missing rate, rate of the observed values which will be artificially masked as missing.
-
-        Note that,
-        `rate` = (number of artificially missing values) / np.sum(~np.isnan(self.data)),
-        not (number of artificially missing values) / np.product(self.data.shape),
-        considering that the given data may already contain missing values,
-        the latter way may be confusing because if the original missing rate >= `rate`,
-        the function will do nothing, i.e. it won't play the role it has to be.
-
-    nan : int/float, optional, default=0
-        Value used to fill NaN values.
-
-    Returns
-    -------
+def little_mcar_test(X):
+    """Little's MCAR Test.
 
+    Refer to :cite:`little1988TestMCAR`
     """
-    # TODO: Create missing values in MNAR case
-    raise NotImplementedError("MNAR case has not been implemented yet.")
+    # TODO: Little's MCAR test
+    raise NotImplementedError("MCAR test has not been implemented yet.")

pycorruptor/mnar.py

@@ -0,0 +1,43 @@
+"""
+Corrupt data by adding missing values to it with MNAR (missing not at random) pattern.
+"""
+
+# Created by Wenjie Du <[email protected]>
+# License: GLP-v3
+
+
+def mnar(X, rate, nan=0):
+    """Create not-random missing values (MNAR case).
+
+    Parameters
+    ----------
+    X : array,
+        Data vector. If X has any missing values, they should be numpy.nan.
+
+    rate : float, in (0,1),
+        Artificially missing rate, rate of the observed values which will be artificially masked as missing.
+
+        Note that,
+        `rate` = (number of artificially missing values) / np.sum(~np.isnan(self.data)),
+        not (number of artificially missing values) / np.product(self.data.shape),
+        considering that the given data may already contain missing values,
+        the latter way may be confusing because if the original missing rate >= `rate`,
+        the function will do nothing, i.e. it won't play the role it has to be.
+
+    nan : int/float, optional, default=0
+        Value used to fill NaN values.
+
+    Returns
+    -------
+
+    """
+    # TODO: Create missing values in MNAR case
+    raise NotImplementedError("MNAR case has not been implemented yet.")
+
+
+def _mnar_numpy(X, rate, nan=0):
+    pass
+
+
+def _mnar_torch(X, rate, nan=0):
+    pass

pycorruptor/utils.py

@@ -0,0 +1,87 @@
+"""
+Utility functions for pycorruptor.
+"""
+
+# Created by Wenjie Du <[email protected]>
+# License: GPL-v3
+
+import numpy as np
+
+try:
+    import torch
+except ImportError:
+    pass
+
+
+def cal_missing_rate(X):
+    """Calculate the originally missing rate of the raw data.
+
+    Parameters
+    ----------
+    X : array-like,
+        Data array that may contain missing values.
+
+    Returns
+    -------
+    originally_missing_rate, float,
+        The originally missing rate of the raw data.
+    """
+    if isinstance(X, list):
+        X = np.asarray(X)
+
+    if isinstance(X, np.ndarray):
+        originally_missing_rate = np.sum(np.isnan(X)) / np.product(X.shape)
+    elif isinstance(X, torch.Tensor):
+        originally_missing_rate = torch.sum(torch.isnan(X)) / np.product(X.shape)
+        originally_missing_rate = originally_missing_rate.item()
+    else:
+        raise TypeError(
+            "X must be type of list/numpy.ndarray/torch.Tensor, " f"but got {type(X)}"
+        )
+
+    return originally_missing_rate
+
+
+def masked_fill(X, mask, val):
+    """Like torch.Tensor.masked_fill(), fill elements in given `X` with `val` where `mask` is True.
+
+    Parameters
+    ----------
+    X : array-like,
+        The data vector.
+
+    mask : array-like,
+        The boolean mask.
+
+    val : float
+        The value to fill in with.
+
+    Returns
+    -------
+    array,
+        mask
+    """
+    assert X.shape == mask.shape, (
+        "Shapes of X and mask must match, "
+        f"but X.shape={X.shape}, mask.shape={mask.shape}"
+    )
+    assert isinstance(X, type(mask)), (
+        "Data types of X and mask must match, " f"but got {type(X)} and {type(mask)}"
+    )
+
+    if isinstance(X, list):
+        X = np.asarray(X)
+        mask = np.asarray(mask)
+
+    if isinstance(X, np.ndarray):
+        mask = mask.astype(bool)
+        X[mask] = val
+    elif isinstance(X, torch.Tensor):
+        mask = mask.type(torch.bool)
+        X[mask] = val
+    else:
+        raise TypeError(
+            "X must be type of list/numpy.ndarray/torch.Tensor, " f"but got {type(X)}"
+        )
+
+    return X