How do I handle “TypeError: add_all_ta_features() got multiple values for keyword argument 'fillna'”?

Question

I'm following this documentation: https://buildmedia.readthedocs.org/media/pdf/technical-analysis-library-in-python/latest/technical-analysis-library-in-python.pdf

Specifically page 9 of the PDF. Copying this code I get a type error. I'm not sure if that means I should edit my data file (which is just standard stock data) or edit the fillna portion of the code.

from ta import *

df = pd.read_csv('VRAY.csv', sep=',')
df = utils.dropna(df)

df = add_all_ta_features(df,"Date","Open","High","Low","Close","Adj_Close","Volume", fillna=True)

This is the error:

Traceback (most recent call last): File "/home/toni/PycharmProjects/PyPractice/stockTA/techanal.py", line 7, in df = add_all_ta_features(df,"Date","Open","High","Low","Close","Adj_Close","Volume", fillna=True) TypeError: add_all_ta_features() got multiple values for keyword argument 'fillna'

Answer 1

The documentation for the ta module gives this example for calling add_all_ta_features():

add_all_ta_features(df, "Open", "High", "Low", "Close", "Volume_BTC", fillna=True)

That call has five columns:

Open
High
Low
Close
Volume_BTC

But your call has seven columns:

Date
Open
High
Low
Close
Adj_Close
Volume

You're trying to use too many columns.

Answer 2

From the project repository, the signature of the add_all_ta_features() method is

def add_all_ta_features(df, open, high, low, close, volume, fillna=False, colprefix=""):

By providing so many arguments in your call, you are assigning a value to fillna positionally as "Adj_Close". You cannot subsequently provide another value as a keyword argument.

Answer 3

You added one more parameter to it... As you wrote extra parameters (9 parameters), one parameter took place of "fillna"'s parameter. So when you used fillna as keyword argument ("fillna=") it was written twice to the same parameter.

It should be written this way (7 parameters):

add_all_ta_features(df, "Open", "High", "Low", "Close", "Volume_BTC", fillna=True)

You can find more examples on ta's GitHub: https://github.com/bukosabino/ta