'DataFrame' object has no attribute 'value_counts'

Question

My dataset is a DataFrame of dimension (840,84). When I write the code: ds[ds.columns[1]].value_counts()
I get a correct output:

Out[82]:
0    847
1      5
Name: o_East, dtype: int64

But when I write a loop to store values, I get 'DataFrame' object has no attribute 'value_counts'. I can't explain why ...

wind_vec = []
wind_vec = [(ds[x].value_counts()) for x in ds.columns]

UPDATE FOR THE CODE

import pandas as pd
import numpy as np
import numpy.ma as ma
import statsmodels.api as sm
import matplotlib
import matplotlib.pyplot as plt
from sklearn.preprocessing import OneHotEncoder

dataset = pd.read_csv('data/dataset.csv')

ds = dataset

o_wdire = pd.get_dummies(ds['o_wdire'])
s_wdire = pd.get_dummies(ds['s_wdire'])
t_wdire = pd.get_dummies(ds['t_wdire'])
k_wdire = pd.get_dummies(ds['k_wdire'])
b_wdire = pd.get_dummies(ds['b_wdire'])

o_wdire.rename(columns={'ENE': 'o_ENE','ESE': 'o_ESE', 'East': 'o_East', 'NE': 'o_NE', 'NNE': 'o_NNE', 'NNW': 'o_NNW', \
                     'NW': 'o_NW', 'North': 'o_North', 'SE': 'o_SE', 'SSE': 'o_SSE', 'SSW': 'o_SSW', 'SW': 'o_SW', \
                     'South': 'o_South', 'Variable': 'o_Variable', 'WSW': 'o_WSW','West':'o_West'}, inplace=True)
s_wdire.rename(columns={'ENE': 's_ENE','ESE': 's_ESE', 'East': 's_East', 'NE': 's_NE', 'NNE': 's_NNE', 'NNW': 's_NNW', \
                     'NW': 's_NW', 'North': 's_North', 'SE': 's_SE', 'SSE': 's_SSE', 'SSW': 's_SSW', 'SW': 's_SW', \
                     'South': 's_South', 'Variable': 's_Variable', 'West': 's_West','WSW': 's_WSW'}, inplace=True)
k_wdire.rename(columns={'ENE': 'k_ENE','ESE': 'k_ESE', 'East': 'k_East', 'NE': 'k_NE', 'NNE': 'k_NNE', 'NNW': 'k_NNW', \
                     'NW': 'k_NW', 'North': 'k_North', 'SE': 'k_SE', 'SSE': 'k_SSE', 'SSW': 'k_SSW', 'SW': 'k_SW', \
                     'South': 'k_South', 'Variable': 'k_Variable', 'WNW': 'k_WNW', 'West': 'k_West','WSW': 'k_WSW'}, inplace=True)
b_wdire.rename(columns={'ENE': 'b_ENE','ESE': 'b_ESE', 'East': 'b_East', 'NE': 'b_NE', 'NNE': 'b_NNE', 'NNW': 'b_NNW', \
                     'NW': 'b_NW', 'North': 'b_North', 'SE': 'b_SE', 'SSE': 'b_SSE', 'SSW': 'b_SSW', 'SW': 'b_SW', \
                     'South': 'b_South', 'Variable': 'b_Variable', 'WSW': 'b_WSW', 'WNW': 'b_WNW', 'West': 'b_West'}, inplace=True)
t_wdire.rename(columns={'ENE': 't_ENE','ESE': 't_ESE', 'East': 't_East', 'NE': 't_NE', 'NNE': 't_NNE', 'NNW': 't_NNW', \
                     'NW': 't_NW', 'North': 't_North', 'SE': 't_SE', 'SSE': 't_SSE', 'SSW': 't_SSW', 'SW': 't_SW', \
                     'South': 't_South', 'Variable': 't_Variable', 'WSW': 't_WSW', 'WNW': 't_WNW', 'West':'t_West'}, inplace=True)

#WIND
ds_wdire = pd.DataFrame(pd.concat([o_wdire,s_wdire,t_wdire,k_wdire,b_wdire],axis=1))
ds_wdire = ds_wdire.astype('float64')

In [93]: ds_wdire.shape
Out[93]: (852, 84)

In[101]: ds_wdire[ds_wdire.columns[0]].head()
Out[101]:   
0    0
1    0
2    0
3    0
4    0
Name: o_ENE, dtype: float64

In[103]: ds_wdire[ds_wdire.columns[0]].value_counts()
Out[103]:

0    838
1     14
Name: o_ENE, dtype: int64


In[104]: [ds_wdire[x].value_counts() for x in ds_wdire.columns]
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-104-d9756c468818> in <module>()
      1 #Filtering for the wind direction based on the most frequent ones.
----> 2 [ds_wdire[x].value_counts() for x in ds_wdire.columns]

<ipython-input-104-d9756c468818> in <listcomp>(.0)
      1 #Filtering for the wind direction based on the most frequent ones.
----> 2 [ds_wdire[x].value_counts() for x in ds_wdire.columns]

/home/florian/anaconda3/lib/python3.5/site-packages/pandas/core/generic.py in __getattr__(self, name)
   2358                 return self[name]
   2359             raise AttributeError("'%s' object has no attribute '%s'" %
-> 2360                                  (type(self).__name__, name))
   2361 
   2362     def __setattr__(self, name, value):

AttributeError: 'DataFrame' object has no attribute 'value_counts'

Solution

Thanks to @EdChum adviced, I checked :

len(ds_wdire.columns),len(ds_wdire.columns.unique())
Out[100]: (83,84)

Actually, there was a missing name value in the dict that should have been modified from 'WNW' to 'o_WNW'.:

o_wdire.rename(columns={'ENE': 'o_ENE','ESE': 'o_ESE', 'East': 'o_East', 'NE': 'o_NE', 'NNE': 'o_NNE', 'NNW': 'o_NNW', \
                 'NW': 'o_NW', 'North': 'o_North', 'SE': 'o_SE', 'SSE': 'o_SSE', 'SSW': 'o_SSW', 'SW': 'o_SW', \
                 'South': 'o_South', 'Variable': 'o_Variable', 'WSW': 'o_WSW','West':'o_West', **[MISSING VALUE WNW]**}, inplace=True)

Maybe it would be better to write a loop that inserts a prefix to the wind direction variables, this way, I would avoid that kind of problem.

Comments

[EdChum]

You can just do df.columns = ['o_' + x for x in df] and similar for other dfs