Get index value from pandas groupby first/last

I am trying to recover the original date or index from grouping a times series with a datetime index by year. Is there a faster way without a loop and an extra column to obtain first_day_indices

import pandas as pd
import numpy as np
import datetime as dt

# Data
T = 1000
base = dt.date.today()
date_list = [base - dt.timedelta(weeks=x) for x in range(T)]
date_list.reverse()
test_data = pd.DataFrame(np.random.randn(T)/100, columns=['Col1'])
test_data.index = pd.to_datetime(date_list)
test_data['date'] = test_data.index

first_days = test_data['date'].groupby(test_data.index.year).first()

first_day_indices= []

for i in first_days:
    first_day_indices.append(np.where(test_data.index == i)[0][0])

print(first_day_indices)

>Solution :

You can use pandas.Series.isin to check whether elements in Series are contained in a list of values.

test_data.reset_index()[test_data.index.isin(first_days)].index.tolist()