Copy a column from one dataframe to another dataframe

[jefsummers]

df1=df does not copy df into a new dataframe df1, despite that being what it looks like. Instead, you simply have 2 variables pointing to the same dataframe. You need to use the copy() command

See Documentation

[deanhystad]

df1 = df does not make a new dataframe. df1 and df are two variables that reference the same dataframe object. I will not go as far as saying assignment never makes a copy, but I will say that I have never seen it.

[Led_Zeppelin]

You are correct and I am wrong. I earned the way t copy a dataframe (deep copy actually) and have it stand alone and not be changed, by any
changes to the original template.

It works perfectly.


However, the final command which was given to me in this series of posts it still throwing errors. The error is:

Error:
ndexError                                Traceback (most recent call last)
Input In [14], in <cell line: 1>()
----> 1 df["machine_status"] = df1["machine_status"]

IndexError: only integers, slices (`:`), ellipsis (`...`), numpy.newaxis (`None`) and integer or boolean arrays are valid indices

I am still not getting the line to work. What is the cause of this error?

Thanks in advance.

Respectfully,

LZ

[deanhystad]

Either df or df1 (or both) is not a DataFrame. What are their types? Looks like one (at least, but guessing df) is a numpy array.

[Led_Zeppelin]

Yes, but how to determine if df or df1 is a dataset?

I am assuming dtype() function will do it. But I do not know the syntax.

Respectfully,

LZ

[deanhystad]

Use the built-in the type() function. print(type(df), type(df1))

[Led_Zeppelin]

You are correct. df is a numpy array. How did you know?

[deanhystad]

It was in the error message:

Error:
IndexError: only integers, slices (`:`), ellipsis (`...`), numpy.newaxis (`None`) and integer or boolean arrays are valid indices

Lots of things can have [integer] or slices, but not many things that aren't numpy will accept a numpy.newaxis as an argument. Even without that I don't know many things that accept "[...]".