merged long-description into the readme; let setup use it as long-des…

…cription

README.md

@@ -1,6 +1,6 @@
 # CplxModule
 
-A lightweight extension for `torch.nn` that adds layers and activations, which respect algebraic operations over the *field of complex numbers*, and implements *real- and complex-valued Variational Dropout* methods for weight sparsification.
+A lightweight extension for `torch.nn` that adds layers and activations, which respect algebraic operations over the *field of complex numbers*, and implements *real- and complex-valued Variational Dropout* methods for weight sparsification. The complex-valued building blocks and Variational Dropout layers of both kinds can be seamlessly integrated into pytorch-based training pipelines. The package provides the toolset necessary to train, sparsify and fine-tune both real- and complex-valued models.
 
 ## Documentation
 

setup.py

@@ -14,12 +14,8 @@
         name="cplxmodule",
         version=version,
         description="Variational Dropout and Complex-valued Neural Networks in pytorch",
-        long_description="""An extension for pytorch providing essential building """
-        """blocks for Complex-valued Neural Networks and `torch.nn`-compatible """
-        """real- and complex-valued Variational Dropout methods. Implements the """
-        """tools necessary to train, sparsify and fine-tune both real- and """
-        """complex-valued models, and seamlessly integrate them into existing """
-        """pipelines.""",
+        long_description=open("README.md").read(),
+        long_description_content_type="text/markdown",
         url="https://github.com/ivannz/cplxmodule",
         keywords=[
             "pytorch",