ID column based on condition in another column

Hej,
I have the following column A which shows continuous data sampling. If the value changes the Sensor is measuring data, if the values repeat itself the sensor is not measuring actual values.
`
I would like to loop through A and assign an ID column with the condition that if the measurement starts and the values are changing continuously the ID should be the same. If the values are not changing the ID should increment by one. If then the values start changing again another increment should be added to the ID column and so forth.
Tried this so far which does not accurately assigns the ID

import pandas as pd


data = [3.5, 3.6, 3.7, 3.8, 1, 1, 1, 1, 1, 3.9, 4.0, 4.2, 4.4, 4.6, 4.8, 3,        
3, 3,  3, 3.2, 3.3, 3.5, 2.1, 2.1, 2.1, 2.1]

df = pd.DataFrame({'A': data})

# Initialize ID 
df['ID'] = 0

# Assign ID
 id_value = 1
 for i in range(1, len(df)):
    if df.at[i, 'A'] != df.at[i-1, 'A']:
       df.at[i, 'ID'] = df.at[i-1, 'ID']
    else:
       df.at[i, 'ID'] = id_value
       id_value += 1

print(df)


Expected output:
      A  ID
 0   3.5   0
 1   3.6   0
 2   3.7   0
 3   3.8   0
 4   1.0   1
 5   1.0   1
 6   1.0   1
 7   1.0   1
 8   1.0   1
 9   3.9   2
 10  4.0   2

…

In my output the data is incremented continuously, so each row has its own ID.

If the expected output is correct you can do this by looking at the value of the previous AND the next row using shift. Assigning the result to a new column so we can make use the cumulative sum against it.

>>> df = pd.DataFrame({'A': data})
>>> df['Changing'] = (df['A'] != df['A'].shift(-1)) & (df['A'] != df['A'].shift())
>>> df
A  Changing
0   3.5      True
1   3.6      True
2   3.7      True
3   3.8      True
4   1.0     False
5   1.0     False
6   1.0     False
7   1.0     False
8   1.0     False
9   3.9      True
10  4.0      True
11  4.2      True
12  4.4      True
13  4.6      True
14  4.8      True
15  3.0     False
16  3.0     False
17  3.0     False
18  3.0     False
19  3.2      True
20  3.3      True
21  3.5      True
22  2.1     False
23  2.1     False
24  2.1     False
25  2.1     False
>>> df['ID'] = (df['Changing'] != df['Changing'].shift()).cumsum() - 1
>>> df
A  Changing  ID
0   3.5      True   0
1   3.6      True   0
2   3.7      True   0
3   3.8      True   0
4   1.0     False   1
5   1.0     False   1
6   1.0     False   1
7   1.0     False   1
8   1.0     False   1
9   3.9      True   2
10  4.0      True   2
11  4.2      True   2
12  4.4      True   2
13  4.6      True   2
14  4.8      True   2
15  3.0     False   3
16  3.0     False   3
17  3.0     False   3
18  3.0     False   3
19  3.2      True   4
20  3.3      True   4
21  3.5      True   4
22  2.1     False   5
23  2.1     False   5
24  2.1     False   5
25  2.1     False   5

I subtracted 1 from the cumsum to follow the expected output. You can then drop the added column as needed.

After several fails, I managed to get the output you wanted

Code

import pandas as pd
data = [3.5, 3.6, 3.7, 3.8, 1, 1, 1, 1, 1, 3.9, 4.0, 4.2, 4.4, 4.6, 4.8, 3, 3, 3, 3, 3.2, 3.3, 3.5, 2.1, 2.1, 2.1, 2.1]
df = pd.DataFrame({'A': data})
# Initialize ID
df['ID'] = 0
# Assign ID
id_value = 0
prev_value = df.at[0, 'A']
value_was_changing = True
for i in range(1, len(df)):
value_is_changing = df.at[i, 'A'] != prev_value
if value_is_changing:
if not value_was_changing:
id_value += 1
else:
if value_was_changing:
id_value += 1
df.at[i, 'ID'] = id_value
prev_value = df.at[i, 'A']
value_was_changing = value_is_changing
print(df)

Output

      A  ID
0   3.5   0
1   3.6   0
2   3.7   0
3   3.8   0
4   1.0   0
5   1.0   1
6   1.0   1
7   1.0   1
8   1.0   1
9   3.9   2
10  4.0   2
11  4.2   2
12  4.4   2
13  4.6   2
14  4.8   2
15  3.0   2
16  3.0   3
17  3.0   3
18  3.0   3
19  3.2   4
20  3.3   4
21  3.5   4
22  2.1   4
23  2.1   5
24  2.1   5
25  2.1   5

Explanation

I edited the script to accurately track whether the values are continuously changing or staying the same, and assign IDs accordingly.

1. Added prev_value variable to keep track of the previous value in the loop
2. Added value_was_changing flag to keep track of whether the previous value was changing or not.
3. Added value_is_changing to determine if the current value is different from the previous value.
4. The loop increments id_value when there's a transition between changing and not changing states.
5. Updated prev_value and value_was_changing at the end of each iteration for use in the next iteration.