Creating a labeled series with labels that correspond to each 'color' and values?

Question

I have an example dataframe (df) like the one shown below, and I would like to use pandas to create a series with labels that correspond to each color and the number of times it appears an entry with that color appears in the dataframe, kind of like a totals for each color. I have tried the following, but Instead get a series with the total number of rows showing as the color sum for each color:

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

df= pd.read_csv('data_set.txt', index _col=0)
total_count = {_:len(df['type']) for _ in df['type'].unique() }
total_count

Current Output:

{'red': 12,
 'green': 12,
 'yellow': 12,
 'blue': 12}

However, clearly there are not 12 entries for each of the 4 colors in the dataframe. What am I doing wrong?

number	date	color	weight	temperature	size
0	1/1/2021	red	0.2	0.2	big
1	1/1/2021	red	0.6	0.6	small
2	1/1/2021	red	0.4	0.6	small
3	1/1/2021	green	0.2	0.4	big
4	1/1/2021	green	1	1	small
5	1/1/2021	yellow	0.4	0.4	big
6	1/1/2021	yellow	0.1	0.2	big
7	1/1/2021	yellow	1.3	0.5	big
8	1/1/2021	yellow	1.5	0.5	small
9	1/1/2021	yellow	1.5	0.5	small
10	1/1/2021	blue	0.4	0.3	big
11	1/1/2021	blue	0.8	0.2	small

Answer 1

try:-

df['color'].value_counts()

Output:-

yellow    5
red       3
green     2
blue      2
Name: color, dtype: int64