Pyodbc doesn't run the procedure correctly without even throwing any error

Question

I have a SQL Server procedure which runs in SQL Server Management Studio (SSMS) without any problem and every time it generates the same table.

I want to run the same procedure using python and pyodbc. In order to do that I wrote a python function run_procedure(my_procedure) which runs the procedure.

The problem is, every time I run the python function, I get a different result!

The procedure:

DROP PROCEDURE IF EXISTS procedure_create_tbl_date_range;
GO

CREATE PROCEDURE procedure_create_tbl_date_range
AS
   DROP TABLE IF EXISTS tbl_date_range;


   DECLARE @dateFrom date;
   DECLARE @dateTo date;

   SET @dateFrom = '2021-12-12';
   SET @dateTo   = '2022-12-12';

   create table tbl_date_range (dates_ date);

   WHILE(@dateFrom < @dateTo)
   BEGIN
      SELECT @dateFrom = DATEADD(day, 1,@dateFrom)
      INSERT INTO tbl_date_range 
      SELECT @dateFrom
   END;

GO

The python function:

import pyodbc
import pandas as pd

def run_procedure(my_procedure):
    conn = pyodbc.connect('Driver={SQL Server};'
                        'Server=.\SQLEXPRESS;'
                        'Database=my_db;'
                        'Trusted_Connection=yes'
                        )

    cur = conn.cursor()
    try:
        res = cur.execute(my_procedure)
        conn.commit()
    except Exception as e:
        print(e)
    finally:
        cur.close()
        del cur
        conn.close()

Now almost each time I run the function below, the query SELECT max(dates_) as max_date_ FROM tbl_date_range returns a different result!

run_procedure("EXEC procedure_create_tbl_date_range")

First try:

Second try:

Does anyone know, why the result is different?!