How to fix invalid literal for int() with base 10 error in pandas: "0.75" in pandas-CodePudding

I am converting data types of dataframe df2 same as df1 by using below code but it is giving me this error.

code:

df2 = df2.astype(df1.dtypes.to_dict())

Error:

invalid literal for int() with base 10 error in pandas: "0.75" in pandas

Is there any general solution to fix this line of code?

I tried converting both the data frames values to string but didn't work.

df2.astype(str) 
df2 = df2.astype(df1.astype(str).dtypes.to_dict())

CodePudding user response：

0.75 is not an integer. Therefore it breaks.

If you want to transform 0.75 to 1 you can round the values and then transform then to integers.

But I guess you should firstly check if thats really what you want to do

CodePudding user response：

I don't know exact situation of your dataset because there is no example.

try this one

cols = df1.select_dtypes('number').columns
df2[cols] = df2[cols].apply(pd.to_numeric, errors='coerce')
df2 = df2.astype(df1.dtypes)

CodePudding user response：

This one might be perfect if you just wanna mirror data type of every column in df1 to df2.

for x in df1.columns:
   df2[x]=df2[x].astype(df1[x].dtypes.name)