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

# In[37]:


import pandas as pd
import numpy as np
from platform import python_version


# In[38]:


python_version(), pd.__version__, np.__version__


# ## loc
# 
# > locate by label

# In[3]:


df_default_index = pd.DataFrame({
    'name':['john','mary','peter','nancy','gary'],
    'age':[22,33,27,22,31],
    'state':['AK','DC','CA','CA','NY']
})
df_default_index


# In[4]:


df_default_index.loc[[0]]


# In[5]:


df_default_index.loc[[2,3]]


# In[6]:


df_name_index = pd.DataFrame(
    index=['john','mary','peter','nancy','gary'],
    data={
        'age':[22,33,27,22,31],
        'state':['AK','DC','CA','CA','NY']
    }
)


# In[7]:


df_name_index


# In[8]:


df_name_index.loc[['peter']]


# In[9]:


df_name_index.loc[['john']]


# ## iloc
# 
# > locate by position

# In[10]:


df_name_index.iloc[[0]]


# In[11]:


df_name_index.iloc[[2,3,4]]


# In[12]:


df_default_index.iloc[[0]]


# In[13]:


df_default_index.iloc[[2,3,4]]


# ## set value to individual  cell
# 
# > must use `loc`

# In[14]:


df_set = pd.DataFrame({
    'name':['john','mary','peter','nancy','gary'],
    'age':[22,33,27,22,31],
    'state':['AK','DC','CA','CA','NY']
})
df_set


# In[15]:


df_set.loc[0,'name'] = 'bartholomew'

df_set.loc[3, 'age'] = 39

df_set


# In[16]:


df_name_index.loc['john','age'] = 99
df_name_index


# ## Use column as index

# In[26]:


df = pd.DataFrame({
    'name':['john','mary','peter','nancy','gary'],
    'age':[22,33,27,22,31],
    'state':['AK','DC','CA','CA','NY']
})
df


# In[30]:


df.set_index('name', verify_integrity=True)


# ## Set multiple values

# In[63]:


df = pd.DataFrame({
    'name':['john','mary','peter','nancy','gary'],
    'age':[22,33,27,22,31],
    'state':['AK','DC','CA','CA','NY'],
    'lives_in_ca': [False,False,False,False,False]
})
df


# In[64]:


index = df[df['state']=='CA'].index


# In[65]:


df.loc[index,'lives_in_ca'] = True
df


# ## settingwithcopywarning

# In[19]:


df = pd.DataFrame({
    'name':['john','mary','peter','nancy','gary'],
    'age':[22,33,27,22,31],
    'state':['AK','DC','CA','CA','NY']
})


# ### bad:

# In[20]:


df_over_30_years = df[df['age']>30]
df_over_30_years


# In[21]:


df_over_30_years['new_column'] = 'some_value'


# ### good:

# In[22]:


df_over_30_years = df.copy()[df['age']>30]
df_over_30_years


# In[23]:


df_over_30_years['new_column'] = 'some_value'


# In[24]:


df_over_30_years


# In[ ]: