Python Errors & Solutions

Events are in reverse chronological order.

• 2020/08/08 - AttributeError: ‘float’ object has no attribute ‘lower’
• 2020/08/08 - ImportError: lxml not found, please install it
• 2020/07/06 - tuple indices must be integers or slices, not str
• 2020/04/17 - TypeError: unorderable types: datetime.date() > str()
• 2020/04/03 - TypeError: ‘tuple’ object is not callable

---

Error : AttributeError: ‘float’ object has no attribute ‘lower’

Error occurs in this line when trying to convert case to lower.

df_tc['name'] = df_tc['name'].apply(lambda x: x.lower())
 
# In df_tc.info(), it confirms its an object but executing above line, it fails
name         3754 non-null object

Why this error occured because NaNs are considered as float ? I got to find that by doing

for row in df_tc.itertuples(): 
    if(type(row.name) is not str):
        print(row.Index, type(row.name))
 
# Output
14 <class 'float'>
21 <class 'float'>
....
 
df_tc.loc[14]
# Output
dt           2020-07-03
name                NaN
cl                 0.76
Name: 14, dtype: object

Solution :

df_tc['name'] = df_tc.name.str.lower()
 

Error : ImportError: lxml not found, please install it

Installing in Jupyter

import sys
!{sys.executable} -m pip install lxml
 
**Output**    
Requirement already satisfied: lxml in c:\users\sushanth\appdata\local\programs\python\python35-32\lib\site-packages (4.5.2)

To know where its installed,

import lxml
lxml.__path__

This error you get even after you have installed everything,

import lxml
import requests
import html5lib
import BeautifulSoup4
 
import sys
!{sys.executable} -m pip install BeautifulSoup4

When you execute the below code,

url = 'https://en.wikipedia.org/wiki/List_of_elevator_accidents'
tables = pd.read_html(url)
 
print(tables)
 
#Output :
ImportError: lxml not found, please install it

Solution :

You have to Restart Kernel in Jupyter

Error : tuple indices must be integers or slices, not str

When looping dataframe trying to get a value from dataframe cell, got the error.

row[rd_Qcolname] 

Solution :

getattr(row, rd_Qcolname)

Error : TypeError: unorderable types: datetime.date() > str()

Date : 17/April/2020

When executing the below line in pandas, got that error.

df_temp[df_temp['ts'] > dt_3d]
 
# Error : 
# TypeError: unorderable types: datetime.date() > str()

Analysis

print(type(df_temp['ts']))    # <class 'pandas.core.series.Series'>
print(type(df_temp['ts'][0])) # <class 'datetime.date'>
print(type(dt_3d))            # <class 'str'>

Update 1 : Updated like below, still got an error.

df_temp[df_temp['ts'] > pd.to_datetime(dt_3d)]
 
# Error : 
# TypeError: Cannot compare type 'Timestamp' with type 'date'

Solution :

Update 2 : df_temp[df_temp['ts'] > pd.to_datetime(dt_3d).date()]

Worked

Error : TypeError: ‘tuple’ object is not callable

for i in range(10): 
    print(i, end =" ") 
 
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-324-e27c647546fc> in <module>
----> 1 for i in range(10):
      2     print(i, end =" ")
 
TypeError: 'tuple' object is not callable

Solution : This took some time to solve. I had used keyword range as a variable.

print(range)
 
(201913, 201915)

Solution is del range and then the loop worked.

for i in range(10): 
    print(i, end =" ") 
 
0 1 2 3 4 5 6 7 8 9