#!/usr/bin/env python
# coding: utf-8

# # Tutorial 3: Data Plots
# 
# This tutorial will talk about how to visualise the distributions that have been built in Tutorial 1. 
# 
# **NOTE FOR CONTRIBUTORS: Always clear all output before commiting (``Cell`` > ``All Output`` > ``Clear``)**!

# In[ ]:


# Magic
get_ipython().run_line_magic('matplotlib', 'inline')
# Reload modules whenever they change
get_ipython().run_line_magic('load_ext', 'autoreload')
get_ipython().run_line_magic('autoreload', '2')

# Make clusterking package available even without installation
import sys
sys.path = ["../../"] + sys.path


# In[ ]:


from clusterking.data.data import Data
from clusterking.plots import BundlePlot


# First we load the data created in Tutorial 1 in the folder output/cluster/ with the name tutorial_basics and pass it to the Data class.

# In[ ]:


d = Data("output/cluster/", "tutorial_basics")


# This data is then used to create an instance of the BundlePlot class.

# In[ ]:


pb = BundlePlot(d)


# We are now ready to visualise our created data: Let's start by drawing the histograms corresponding to the benchmark points of each clusters by typing:

# In[ ]:


pb.plot_bundles()


# We can also add more sample points to the plot (in addition to the benchmark point):

# In[ ]:


pb.plot_bundles(1, nlines=3)


# To save the above plot to the output/cluster folder we use the following commnad:

# In[ ]:


pb.fig.savefig("output/cluster/example_plot.png")


# Showing the minima and maxima of all clusters is achieved with the plot_minmax method. 

# In[ ]:


pb.plot_minmax()


# The same plot for clusters 2 and 3 only:

# In[ ]:


pb.plot_minmax([2,3])


# Removing the reference line leads to the follwoing output:

# In[ ]:


pb.plot_minmax([2,  3], reference=False)


# Box plots can be produced using  the box_plot method:

# In[ ]:


pb.box_plot(reference=True)


# Showing clusters 0 and 2 only:

# In[ ]:


pb.box_plot([0, 2])