Updates example section in the Readme (#48)

* removes jupyter artifacts
* adds extended description of the fruits example and clarifies use for larger datasets
* updates status badge and gives ci workflow the name build to match with travis
* moved plot creation script to .github as it has nothing to do with the package and is only used for the repo
* simplifies data aggregation in the preprocessing example
* removes old png file from package folder (the ones still used can be found in the .github/ folder

Co-authored-by: Pierre Sassoulas <pierre.sassoulas@gmail.com>

.github/create_documentation_plots.py

@@ -0,0 +1,53 @@
+import matplotlib.pyplot as plt
+import pandas as pd
+from pysankey import sankey
+
+df = pd.read_csv("../pysankey/fruits.txt", sep=" ", names=["true", "predicted"])
+
+colorDict = {
+    "apple": "#f71b1b",
+    "blueberry": "#1b7ef7",
+    "banana": "#f3f71b",
+    "lime": "#12e23f",
+    "orange": "#f78c1b",
+    "kiwi": "#9BD937",
+}
+
+labels = list(colorDict.keys())
+leftLabels = [label for label in labels if label in df["true"].values]
+rightLabels = [label for label in labels if label in df["predicted"].values]
+
+ax = sankey(
+    left=df["true"],
+    right=df["predicted"],
+    leftLabels=leftLabels,
+    rightLabels=rightLabels,
+    colorDict=colorDict,
+    aspect=20,
+    fontsize=12,
+)
+
+plt.savefig("img/fruits.png")
+plt.close()
+
+
+# This calculates how often the different combinations of "true" and
+# "predicted" co-occure
+df = df.groupby(["true", "predicted"]).size().reset_index()
+weights = df[0].astype(float)
+
+
+ax = sankey(
+    left=df["true"],
+    right=df["predicted"],
+    rightWeight=weights,
+    leftWeight=weights,
+    leftLabels=leftLabels,
+    rightLabels=rightLabels,
+    colorDict=colorDict,
+    aspect=20,
+    fontsize=12,
+)
+
+
+plt.savefig("img/fruits_weighted.png")

.github/workflows/ci.yaml

@@ -1,3 +1,5 @@
+name: build
+
 on:
   push:
     branches:

.gitignore

@@ -1,3 +1,5 @@
+# Jupyter
+.ipynb_checkpoints/
 
 #Eclipse/pydev
 .project

README.md

@@ -4,14 +4,17 @@ Uses matplotlib to create simple <a href="https://en.wikipedia.org/wiki/Sankey_d
 Sankey diagrams</a> flowing only from left to right.
 
 [![PyPI version](https://badge.fury.io/py/pySankeyBeta.svg)](https://badge.fury.io/py/pySankeyBeta)
-[![Build Status](https://travis-ci.org/Pierre-Sassoulas/pySankey.svg?branch=master)](https://travis-ci.org/Pierre-Sassoulas/pySankey)
+[![Build Status](https://github.com/Pierre-Sassoulas/pySankey/actions/workflows/ci.yaml/badge.svg)](https://github.com/Pierre-Sassoulas/pySankey/actions/workflows/ci.yaml)
 [![Coverage Status](https://coveralls.io/repos/github/Pierre-Sassoulas/pySankey/badge.svg?branch=master)](https://coveralls.io/github/Pierre-Sassoulas/pySankey?branch=master)
 [![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/ambv/black)
 [![License: GPL v3](https://img.shields.io/badge/License-GPLv3-blue.svg)](https://www.gnu.org/licenses/gpl-3.0)
 
-## Example
+## Examples
 
-With fruits.txt :
+### Simple expected/predicted example with fruits.txt:
+
+`pysankey` contains a simple expected/predicted dataset called `fruits.txt` which looks
+like the following:
 
 <div>
 <table border="1" class="dataframe">
@@ -80,9 +83,13 @@ import pandas as pd
 from pysankey import sankey
 import matplotlib.pyplot as plt
 
+
 df = pd.read_csv(
-    'pysankey/fruits.txt', sep=' ', names=['true', 'predicted']
+    'fruits.txt',
+    sep=' ',
+    names=['true', 'predicted']
 )
+
 colorDict = {
     'apple':'#f71b1b',
     'blueberry':'#1b7ef7',
@@ -92,83 +99,77 @@ colorDict = {
     'kiwi':'#9BD937'
 }
 
+labels = list(colorDict.keys())
+leftLabels = [label for label in labels if label in df['true'].values]
+rightLabels = [label for label in labels if label in df['predicted'].values]
+
+# Create the sankey diagram
 ax = sankey(
-    df['true'], df['predicted'], aspect=20, colorDict=colorDict,
-    leftLabels=['banana','orange','blueberry','apple','lime'],
-    rightLabels=['orange','banana','blueberry','apple','lime','kiwi'],
+    left=df['true'],
+    right=df['predicted'],
+    leftLabels=leftLabels,
+    rightLabels=rightLabels,
+    colorDict=colorDict,
+    aspect=20,
     fontsize=12
 )
+
 plt.show() # to display
-plt.savefig('fruit.png', bbox_inches='tight') # to save
 ```
 
-![Fruity Alchemy](pysankey/fruit.png)
+![Fruity Alchemy](.github/img/fruits.png)
 
-You could also use weight:
+### Plotting preprocessed data using weights
 
-```
-,customer,good,revenue
-0,John,fruit,5.5
-1,Mike,meat,11.0
-2,Betty,drinks,7.0
-3,Ben,fruit,4.0
-4,Betty,bread,2.0
-5,John,bread,2.5
-6,John,drinks,8.0
-7,Ben,bread,2.0
-8,Mike,bread,3.5
-9,John,meat,13.0
-```
+However, the data may not always be available in the format mentioned in the previous
+example (for instance, if the dataset is too large). In such cases, the weights between
+the true and predicted labels can be calculated in advance and used to create the Sankey
+diagram. In this example, we will continue working with the data that was loaded in the
+previous example:
 
 ```python
-import pandas as pd
-from pysankey import sankey
-import matplotlib.pyplot as plt
-
-df = pd.read_csv(
-    'pysankey/customers-goods.csv', sep=',',
-    names=['id', 'customer', 'good', 'revenue']
-)
-weight = df['revenue'].values[1:].astype(float)
+# Calculate the weights from the fruits dataframe
+df = df.groupby(["true", "predicted"]).size().reset_index()
+weights = df[0].astype(float)
 
 ax = sankey(
-      left=df['customer'].values[1:], right=df['good'].values[1:],
-      rightWeight=weight, leftWeight=weight, aspect=20, fontsize=20
+    left=df['true'],
+    right=df['predicted'],
+    rightWeight=weights,
+    leftWeight=weights,
+    leftLabels=leftLabels,
+    rightLabels=rightLabels,
+    colorDict=colorDict,
+    aspect=20,
+    fontsize=12
 )
+
 plt.show() # to display
-plt.savefig('customers-goods.png', bbox_inches='tight') # to save
 ```
 
-![Customer goods](pysankey/customers-goods.png)
-
-Similar to seaborn, you can pass a matplotlib `Axes` to `sankey` function:
-
-```python
-import pandas as pd
-from pysankey import sankey
-import matplotlib.pyplot as plt
-
-df = pd.read_csv(
-        'pysankey/fruits.txt',
-        sep=' ', names=['true', 'predicted']
-)
-colorDict = {
-    'apple': '#f71b1b',
-    'blueberry': '#1b7ef7',
-    'banana': '#f3f71b',
-    'lime': '#12e23f',
-    'orange': '#f78c1b'
-}
-
-ax1 = plt.axes()
-
-sankey(
-      df['true'], df['predicted'], aspect=20, colorDict=colorDict,
-      fontsize=12, ax=ax1
-)
-
-plt.show()
-```
+![Fruity Alchemy](.github/img/fruits_weighted.png)
+
+### pysankey function overview
+
+> `sankey(left, right, leftWeight=None, rightWeight=None, colorDict=None, leftLabels=None, rightLabels=None, aspect=4, rightColor=False, fontsize=14, ax=None, color_gradient=False, alphaDict=None)`
+>
+> **left**, **right** : NumPy array of object labels on the left and right of the
+> diagram
+>
+> **leftWeight**, **rightWeight** : Numpy arrays of the weights each strip
+>
+> **colorDict** : Dictionary of colors to use for each label
+>
+> **leftLabels**, **rightLabels** : order of the left and right labels in the diagram
+>
+> **aspect** : vertical extent of the diagram in units of horizontal extent
+>
+> **rightColor** : If true, each strip in the diagram will be be colored according to
+> its left label
+>
+> **fontsize** : Fontsize to be used for the labels
+>
+> **ax** : matplotlib axes to plot on, otherwise uses current axes.
 
 ## Important informations