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

# ![Pandas Tutorial | Hedaro >](https://www.dropbox.com/s/220ncn0o5danuey/pandas-ipython-tutorials-hedaro.jpg?dl=1)
# # Lesson 4

# In this lesson were going to go back to the basics. We will be working with a small data set so that you can easily understand what I am trying to explain. We will be adding columns, deleting columns, and slicing the data many different ways. Enjoy!

# In[1]:


# Import libraries
import pandas as pd
import sys


# In[2]:


print('Python version ' + sys.version)
print('Pandas version: ' + pd.__version__)


# In[3]:


# Our small data set
d = [0,1,2,3,4,5,6,7,8,9]

# Create dataframe
df = pd.DataFrame(d)
df


# In[4]:


# Lets change the name of the column
df.columns = ['Rev']
df


# In[5]:


# Lets add a column
df['NewCol'] = 5
df


# In[6]:


# Lets modify our new column
df['NewCol'] = df['NewCol'] + 1
df


# In[7]:


# We can delete columns
del df['NewCol']
df


# In[8]:


# Lets add a couple of columns
df['test'] = 3
df['col'] = df['Rev']
df


# In[9]:


# If we wanted, we could change the name of the index
i = ['a','b','c','d','e','f','g','h','i','j']
df.index = i
df


# We can now start to select pieces of the dataframe using ***loc***.

# In[10]:


df.loc['a']


# In[11]:


# df.loc[inclusive:inclusive]
df.loc['a':'d']


# In[12]:


# df.iloc[inclusive:exclusive]
# Note: .iloc is strictly integer position based. It is available from [version 0.11.0] (http://pandas.pydata.org/pandas-docs/stable/whatsnew.html#v0-11-0-april-22-2013) 
df.iloc[0:3]


# We can also select using the column name.

# In[13]:


df['Rev']


# In[14]:


df[['Rev', 'test']]


# In[15]:


# df.ix[rows,columns]
# replaces the deprecated ix function
#df.ix[0:3,'Rev']
df.loc[df.index[0:3],'Rev']


# In[16]:


# replaces the deprecated ix function
#df.ix[5:,'col']
df.loc[df.index[5:],'col']


# In[17]:


# replaces the deprecated ix function
#df.ix[:3,['col', 'test']]
df.loc[df.index[:3],['col', 'test']]


# There is also some handy function to select the top and bottom records of a dataframe.

# In[18]:


# Select top N number of records (default = 5)
df.head()


# In[19]:


# Select bottom N number of records (default = 5)
df.tail()


# <p class="text-muted">This tutorial was created by <a href="http://www.hedaro.com" target="_blank"><strong>HEDARO</strong></a></p>