# coding: utf-8

# In[1]:


from matplotlib import pyplot as plt


# In[2]:


import pandas as pd


# In[4]:


species = pd.read_csv('species_info.csv')


# In[10]:


species.head()


# In[37]:


species.info()


# In[40]:


species_type = species.category.unique()
print species_type


# In[39]:


conservation_info = species.conservation_status.unique()
print conservation_info


# In[41]:


conservation_num = species.groupby('conservation_status').scientific_name.count().reset_index()
print conservation_num


# In[47]:


species.fillna('No Intervention', inplace=True)


# In[48]:


conservation_num_np = species.groupby('conservation_status').scientific_name.count().reset_index()
print conservation_num_np


# In[90]:


protection_counts = species.groupby('conservation_status')    .scientific_name.nunique().reset_index()    .sort_values(by='scientific_name')
print protection_counts


# In[51]:


plt.figure(figsize=(10, 4))
ax = plt.subplot()
plt.bar(range(len(protection_counts)),
protection_counts.scientific_name.values)
ax.set_xticks(range(len(protection_counts)))
ax.set_xticklabels(protection_counts.conservation_status.values)
plt.ylabel('Number of Species')
plt.title('Conservation Status by Species')
labels = [e.get_text() for e in ax.get_xticklabels()]
print ax.get_title()
plt.show()


# In[89]:


species['is_protected'] = species.conservation_status != 'No Intervention'


# In[91]:


category_counts = species.groupby(['category', 'is_protected']).scientific_name.nunique().reset_index()
category_counts.head(7)


# In[60]:


category_pivot = category_counts.pivot(columns='is_protected', index='category', values='scientific_name').reset_index()
category_pivot.head()


# In[103]:


category_pivot.columns = ['category', 'not_protected', 'protected']
print category_pivot.columns