update readme and docs

README.md

@@ -313,7 +313,7 @@ However, you cannot define your own target types, as that would require re-imple
 
 ### A List of Implemented Transformations
 
-Point transformations:
+Intensity-based point transformations:
 ```python
 Normalize
 NormalizeMeanStd
@@ -324,14 +324,14 @@ RandomBrightnessContrast
 RandomGamma
 ```
 
-Local transformations:
+Intensity-based local transformations:
 ```python
 GaussianBlur 
 RandomGaussianBlur
 RemoveBackgroundGaussian
 ```
 
-Geometric transformations:
+Geometric (spatial) transformations:
 ```python
 AffineTransform
 Resize 
@@ -347,6 +347,14 @@ RandomFlip
 RandomCrop
 ```
 
+Other transformations:
+```python
+Contiguous
+StandardizeDatatype
+ConversionToFormat
+```
+
+
 ### Runtime
 
 Here, we present the execution times (mean ± standard deviation) of individual transformations from our library 

docs/source/examples.rst

@@ -133,8 +133,8 @@ images of shape ``[150, 300, 300]``, ``[1, 150, 300, 300]``, as well as ``[4, 15
 If you want to use a multi-channel :class:`mask` or :class:`float_mask`, you have to split it into
 a set of single-channel :class:`mask` or :class:`float_mask` targets, respectively, and input them
 as stand-alone targets (see
-`the respective section below <https://biovolumentations.readthedocs.io/1.3.1/examples.html#example-transforming-multiple-targets-of-the-same-type>`_.
-below on transforming multiple masks with a single image).
+`the respective section below <https://biovolumentations.readthedocs.io/1.3.1/examples.html#example-transforming-multiple-targets-of-the-same-type>`_
+on transforming multiple masks with a single image).
 
 The :class:`keypoints` target is represented as a list of tuples. Each tuple represents
 the absolute coordinates of a keypoint in the volume, so it must contain either 3 or 4 numbers
@@ -148,6 +148,9 @@ as keyword arguments of a call to the :class:`Compose` object. Consequently, the
 dictionary using the same keys. The default key values are :class:`'image'`, :class:`'mask'`, :class:`'float_mask'`,
 :class:`'keypoints'`, :class:`'bboxes'`, and :class:`'value'`.
 
+Prior to applying any user-defined transformation, the :class:`mask` and :class:`float_mask` targets are type-casted to
+integer and floating-point datatypes, respectively.
+
 Importantly, there must always be an :class:`image`-type target in the sample.
 
 You cannot define your own target types; that would require re-implementing all existing transforms.