I have defined a simple function whose arguments include a list of column names from a dataframe. The intention of the function is
Currently the function only returns the first item in the list successfully and does not loop over the whole list in the args.
I have tried simplifying the function back to copying the columns from one DF and appending these as columns to the new dataframe, however the loop still breaks after the first item.
def gen_signal(self, model_outputs: list, start_date: str, end_date: str, mean = 0):
signal_df = pd.DataFrame()
for model in model_outputs:
model_preds_df = self.model_preds_df.set_index(self.model_preds_df['dates'])
model_result = pd.Series(self.model_preds_df[model],
index = model_preds_df.index)
filter_by_date_params = (model_result.index > start_date) & \
(model_result.index < end_date)
results_data = self.model_preds_df[model][filter_by_date_params]
#perform some logic that has been omitted here for brevity
signal_final = np.mean(results_data)
signal_df[model + '_signal' + '_{0}'.format(start_date)] = signal_final
return signal_df
# test function
per_start = '2002-01-01'
per_end = '2019-07-01'
# call function passing list of df columns to perform op over
gen_signal(['model_1', 'model_2', 'model_3'], per_start, per_end)
#function returns signal df but for only one item in the list.
dates | model_1_signal
_______| ________________
Below is a quick demo of the bug you are facing: (As mentioned by emilanov your return statement is part of the loop and should be moved out)
lst = [1,2,3]
def calc_sum_with_bug(lst):
s = 0
for x in lst:
s += x
return s
def calc_sum(lst):
s = 0
for x in lst:
s += x
return s
print(calc_sum_with_bug(lst))
print(calc_sum(lst))
output
1
6
Collected from the Internet
Please contact [email protected] to delete if infringement.
Comments