In [1]:

    

import pandas as pd
import numpy as np


    

In [2]:

    

df_simple = pd.DataFrame(np.arange(12).reshape(3, 4))


    

In [3]:

    

print(df_simple)


    

    




   0  1   2   3
0  0  1   2   3
1  4  5   6   7
2  8  9  10  11

In [4]:

    

print(df_simple.values)


    

    




[[ 0  1  2  3]
 [ 4  5  6  7]
 [ 8  9 10 11]]

In [5]:

    

print(type(df_simple.values))


    

    




<class 'numpy.ndarray'>

In [6]:

    

print(df_simple.columns)


    

    




RangeIndex(start=0, stop=4, step=1)

In [7]:

    

print(type(df_simple.columns))


    

    




<class 'pandas.core.indexes.range.RangeIndex'>

In [8]:

    

print(df_simple.index)


    

    




RangeIndex(start=0, stop=3, step=1)

In [9]:

    

print(type(df_simple.index))


    

    




<class 'pandas.core.indexes.range.RangeIndex'>

In [10]:

    

print(list(df_simple.columns))


    

    




[0, 1, 2, 3]

In [11]:

    

print(type(list(df_simple.columns)))


    

    




<class 'list'>

In [12]:

    

print(df_simple.columns.tolist())


    

    




[0, 1, 2, 3]

In [13]:

    

print(type(df_simple.columns.tolist()))


    

    




<class 'list'>

In [14]:

    

df = pd.DataFrame(np.arange(12).reshape(3, 4),
                  columns=['col_0', 'col_1', 'col_2', 'col_3'],
                  index=['row_0', 'row_1', 'row_2'])


    

In [15]:

    

print(df)


    

    




       col_0  col_1  col_2  col_3
row_0      0      1      2      3
row_1      4      5      6      7
row_2      8      9     10     11

In [16]:

    

print(df.columns)


    

    




Index(['col_0', 'col_1', 'col_2', 'col_3'], dtype='object')

In [17]:

    

print(type(df.columns))


    

    




<class 'pandas.core.indexes.base.Index'>

In [18]:

    

print(df.index)


    

    




Index(['row_0', 'row_1', 'row_2'], dtype='object')

In [19]:

    

print(type(df.index))


    

    




<class 'pandas.core.indexes.base.Index'>

In [20]:

    

print(df.columns.tolist())


    

    




['col_0', 'col_1', 'col_2', 'col_3']

In [21]:

    

print(type(df.columns.tolist()))


    

    




<class 'list'>

In [22]:

    

print(df.columns[0])


    

    




col_0

In [23]:

    

# df.columns[0] = 'Col_0'
# TypeError: Index does not support mutable operations


    

In [24]:

    

df.columns = ['Col_0', 'Col_1', 'Col_2', 'Col_3']


    

In [25]:

    

df.index = ['Row_0', 'Row_1', 'Row_2']


    

In [26]:

    

print(df)


    

    




       Col_0  Col_1  Col_2  Col_3
Row_0      0      1      2      3
Row_1      4      5      6      7
Row_2      8      9     10     11

In [27]:

    

# df.values = np.arange(12).reshape(3, 4) * 10
# AttributeError: can't set attribute


    

In [28]:

    

print(df['Col_1'])


    

    




Row_0    1
Row_1    5
Row_2    9
Name: Col_1, dtype: int64

In [29]:

    

print(type(df['Col_1']))


    

    




<class 'pandas.core.series.Series'>

In [30]:

    

print(df.Col_1)


    

    




Row_0    1
Row_1    5
Row_2    9
Name: Col_1, dtype: int64

In [31]:

    

print(df.loc[:, 'Col_1'])


    

    




Row_0    1
Row_1    5
Row_2    9
Name: Col_1, dtype: int64

In [32]:

    

print(df.loc['Row_1', :])


    

    




Col_0    4
Col_1    5
Col_2    6
Col_3    7
Name: Row_1, dtype: int64

In [33]:

    

print(type(df.loc['Row_1', :]))


    

    




<class 'pandas.core.series.Series'>

In [34]:

    

print(df.loc['Row_1'])


    

    




Col_0    4
Col_1    5
Col_2    6
Col_3    7
Name: Row_1, dtype: int64

In [35]:

    

print(df.loc[['Row_0', 'Row_2'], ['Col_1', 'Col_3']])


    

    




       Col_1  Col_3
Row_0      1      3
Row_2      9     11

In [36]:

    

print(type(df.loc[['Row_0', 'Row_2'], ['Col_1', 'Col_3']]))


    

    




<class 'pandas.core.frame.DataFrame'>

In [37]:

    

print(df.loc['Row_0', 'Col_1'])


    

    




1

In [38]:

    

print(type(df.loc['Row_0', 'Col_1']))


    

    




<class 'numpy.int64'>

In [39]:

    

print(df.at['Row_0', 'Col_1'])


    

    




1

In [40]:

    

# print(df.at[:, 'Col_1'])
# TypeError: 'slice(None, None, None)' is an invalid key


    

In [41]:

    

print(df.iloc[[0, 2], [1, 3]])


    

    




       Col_1  Col_3
Row_0      1      3
Row_2      9     11

In [42]:

    

print(df.iat[0, 1])


    

    




1

In [43]:

    

print(df.query('Col_0 > 2'))


    

    




       Col_0  Col_1  Col_2  Col_3
Row_1      4      5      6      7
Row_2      8      9     10     11

In [44]:

    

df.loc[:, 'Col_1'] = [10, 50, 90]


    

In [45]:

    

print(df)


    

    




       Col_0  Col_1  Col_2  Col_3
Row_0      0     10      2      3
Row_1      4     50      6      7
Row_2      8     90     10     11

In [46]:

    

df.loc[:] = np.arange(12).reshape(3, 4) * 100


    

In [47]:

    

print(df)


    

    




       Col_0  Col_1  Col_2  Col_3
Row_0      0    100    200    300
Row_1    400    500    600    700
Row_2    800    900   1000   1100

In [48]:

    

# df.loc[:, 'Col_1'] = [10, 50, 90, 130]
# ValueError: Length of values does not match length of index


    

In [49]:

    

# df.loc[:] = np.arange(16).reshape(4, 4) * 100
# ValueError: cannot set using a slice indexer with a different length than the value


    

In [50]:

    

df_multi = pd.read_csv('data/src/sample_pandas_normal.csv', index_col=0)


    

In [51]:

    

print(df_multi)


    

    




         age state  point
name                     
Alice     24    NY     64
Bob       42    CA     92
Charlie   18    CA     70
Dave      68    TX     70
Ellen     24    CA     88
Frank     30    NY     57

In [52]:

    

print(df_multi.dtypes)


    

    




age       int64
state    object
point     int64
dtype: object