I have a list(region_props_list), size = 37, which has the values of 2D numpy arrays like below. (So, region_props_list[0] is a numpy array.)
I want to vertically stack all the data and make it as a new pandas DataFrame, which has a shape of ($$, 7)
How can I vertically stack the data with a for loop? can someone give me an advice?
You don't need a for loop. You can use np.vstack instead:
import numpy as np
lst = [np.array([[1, 2], [3, 4]]), np.array([[5, 6]]), np.array([[7, 8], [9, 10]])]
a = np.vstack(lst)
print(a)
# [[ 1 2]
# [ 3 4]
# [ 5 6]
# [ 7 8]
# [ 9 10]]
If your goal is to construct a dataframe, then you can use itertools.chain with pd.DataFrame.from_records (without even making the v-stacked array):
import numpy as np
import pandas as pd
import itertools
lst = [np.array([[1, 2], [3, 4]]), np.array([[5, 6]]), np.array([[7, 8], [9, 10]])]
df = pd.DataFrame.from_records(itertools.chain.from_iterable(lst))
print(df)
# 0 1
# 0 1 2
# 1 3 4
# 2 5 6
# 3 7 8
# 4 9 10
P.S. Please don't post a screenshot. Make a copy & paste-able minimal example which people can easily work on.
for loop is better. On the contrary, I think you should avoid a for loop if possible, when a list is long. You can use pd.DataFrame.from_records directly to make a dataframe, with itertools.chain. In this way you don't even need to make the intermediate stacked array. Please check the updated answer.
DataFrame is given a suitable array, it will use that without making a copy. You might want to time test your alternatives.