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

# In[301]:


get_ipython().run_line_magic('matplotlib', 'inline')

import matplotlib
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd


# In[302]:


inst = pd.read_csv('instructors.csv', names=['date','cumulative instructors trained'], parse_dates=['date'])
work = pd.read_csv('workshops.csv', names=['date', 'cumulative workshops run'], parse_dates=['date'])
lern = pd.read_csv('learners.csv', names=['date','cumulative learners reached'], parse_dates=['date'])


# In[303]:


full = pd.merge(work, lern, on='date')
full['learners reached'] = full['cumulative learners reached'].diff().fillna(full['cumulative learners reached'])
full['workshops run'] = full['cumulative workshops run'].diff().fillna(full['cumulative workshops run'])
full['workshop size'] = full['learners reached']/full['workshops run']
full['date_int'] = full.date.astype(np.int64)

full


# In[326]:


ax1 = lern.plot(x='date', figsize=(10.0,10.0), title="Software Carpentry Growth", kind='area', stacked=False, alpha=0.5, color='g')
ax2 = ax1.twinx()
ax2 = work.plot(x='date', color='r', ax=ax2, kind='area', stacked=False, alpha=0.5)
ax2.legend(loc=0)
ax1.legend(loc = (0.05,0.9))
ax2.legend(loc = (0.5,0.9))
 
for tl in ax1.get_yticklabels():
    tl.set_color('g')
for tl in ax2.get_yticklabels():
    tl.set_color('r')


# In[333]:


ax1 = inst.plot(x='date', figsize=(10.0,10.0), title="Software Carpentry Growth", kind='area', stacked=False, alpha=0.5, color='b')


# In[334]:


ax1 = inst.plot(x='date', figsize=(10.0,10.0), title="Software Carpentry Growth", kind='area', stacked=False, alpha=0.5, color='b')
ax2 = ax1.twinx()
ax2 = work.plot(x='date', color='r', ax=ax2, kind='area', stacked=False, alpha=0.5)
ax2.legend(loc=0)
ax1.legend(loc = (0.05,0.9))
ax2.legend(loc = (0.5,0.9))
 
for tl in ax1.get_yticklabels():
    tl.set_color('g')
for tl in ax2.get_yticklabels():
    tl.set_color('r')


# In[305]:


# Workshops and learners, discrete
#ax1 = full.plot(x='date', y='learners reached', figsize=(10.0,10.0), title="Software Carpentry Growth")
ax1 = full.plot(x='date', y='workshop size', figsize=(10.0,10.0), title="Software Carpentry Growth", kind='area', stacked=False, alpha=0.2)

ax2 = ax1.twinx()
ax2 = full.plot(x='date', y='workshops run', color='r', ax=ax2, ylim=(0.0,100.0), kind='area', stacked=False, alpha=0.5)
#ax2 = full.plot(x='date', y='workshop size', color='g', ax=ax2, ylim=(0.0,100.0))

ax2.legend(loc=0)
ax1.legend(loc = (0.05,0.95))
ax2.legend(loc = (0.7,0.95))
 
for tl in ax1.get_yticklabels():
    tl.set_color('b')
for tl in ax2.get_yticklabels():
    tl.set_color('r')


# In[306]:


pd.to_datetime(full['date'][:]).month


# In[346]:


# Workshops and learners, discrete
#ax1 = full.plot(x='date', y='learners reached', figsize=(10.0,10.0), title="Software Carpentry Growth")
ax1.clear()
ax1 = full.plot(kind='scatter',x='date_int',y='workshops run', s=full['workshop size']*10, figsize=(5.0,5.0), title="Software Carpentry Workshop Sizes", alpha=0.2)
ax1.legend(loc = (0.05,0.95))

for tl in ax1.get_yticklabels():
    tl.set_color('b')



# In[345]:


ax1 = inst.plot(x='date')
ax2 = ax1.twinx()
ax2 = work.plot(x='date', color='g', ax=ax2)
ax1.legend(loc = (0.05,0.85))
ax2.legend(loc = (0.05,0.7))

for tl in ax1.get_yticklabels():
    tl.set_color('b')

for t2 in ax2.get_yticklabels():
    t2.set_color('g')



# In[335]:


full


# In[ ]: