make ready for pypi

README.md

@@ -2,6 +2,7 @@
 
 [![Documentation Status](https://readthedocs.org/projects/firelight/badge/?version=latest)](https://firelight.readthedocs.io/en/latest/?badge=latest)
 [![Conda Version](https://img.shields.io/conda/vn/conda-forge/firelight.svg)](https://anaconda.org/conda-forge/firelight)
+[![PyPI version](https://badge.fury.io/py/firelight.svg)](https://badge.fury.io/py/firelight)
 
 Firelight is a visualization library for pytorch. 
 Its core object is a **visualizer**, which can be called passing some states (such as `inputs`, `target`, 
@@ -20,7 +21,7 @@ Why you will keep using firelight:
 
 ## Installation
 
-### From source (recommended to get the newest version)
+### From source (to get the most recent version)
 On python 3.6+:
 
 ```bash
@@ -30,13 +31,20 @@ cd firelight/
 # Install
 python setup.py install
 ```
-### From conda
+### Using conda
 
 Firelight is available on conda-forge for python > 3.6 and all operating systems:
 ```bash
 conda install -c pytorch -c conda-forge firelight
 ```
 
+### Using pip
+
+In an environment with [scikit-learn](https://scikit-learn.org/stable/install.html) installed:
+```bash
+pip install firelight
+```
+
 ## Example
 
 - Run the example `firelight/examples/example_data.py`
@@ -111,7 +119,7 @@ plt.imsave('visualizations/example_visualization.jpg', image_grid.numpy())
 
 Resulting visualization: 
 
-![Example Image Grid](examples/example_visualization.png)
+![Example Image Grid](https://raw.githubusercontent.com/inferno-pytorch/firelight/master/examples/example_visualization.png)
 
 Many more visualizers are available. Have a look at [visualizers.py](/firelight/visualizers/visualizers.py ) and [container_visualizers.py](/firelight/visualizers/container_visualizers.py) or, for a more condensed list, the imports in [config_parsing.py](/firelight/config_parsing.py).
 

firelight/__init__.py

@@ -1,3 +1,3 @@
 from .config_parsing import get_visualizer
 
-__version__ = '0.2.0'
+__version__ = '0.2.1'

setup.py

@@ -3,14 +3,21 @@
 """
 
 import setuptools
+import os
 
+def read_file(filename):
+    with open(os.path.join(os.path.dirname(__file__), filename)) as f:
+        return f.read()
 
 setuptools.setup(
     name="firelight",
     author="Roman Remme",
     author_email="roman.remme@iwr.uni-heidelberg.de",
     description="A visualization library for PyTorch tensors.",
-    version="0.2.0",
+    long_description=read_file('README.md'),
+    long_description_content_type='text/markdown',
+    url='https://github.com/inferno-pytorch/firelight',
+    version="0.2.1",
     install_requires=[
         "pyyaml>=3.12",
         "matplotlib",
@@ -20,8 +27,13 @@
         "torch",
     ],
     extras_requires={
-    'umap': ['umap-learn>=0.3.8'],
+        'umap': ['umap-learn>=0.3.8'],
     },
     license="Apache Software License 2.0",
     packages=setuptools.find_packages(),
+    classifiers=[
+        "Intended Audience :: Science/Research",
+        "Programming Language :: Python :: 3",
+        "License :: OSI Approved :: Apache Software License",
+    ]
 )