Python, Loop to reading in Excel file sheets, change header row number

I have a loop that counts the rows in each sheet of an xls. When I open the xls itself the count is not aligning with what python is returning me.

It is due to the first sheet header being in row 3. How can I alter my code to read the first sheet ONLY in at row 3 and ignore the first two lines? The rest of my sheets ALWAYS start at the top row and contain no header. I would like to count the len of my first sheet without header included.

However when I open up my excel and count my sheet I am getting

65522 , header starts in row 3, expecting a count of 65520
65520
65520
65520
65520
65520
65520
65520
65520
65520
65520
25427

my full code:

from io import BytesIO
from pathlib import Path
from zipfile import ZipFile
import os
import pandas as pd
from os import walk


def process_files(files: list) -> pd.DataFrame:
    file_mapping = {}
    for file in files:
        #data_mapping = pd.read_excel(BytesIO(ZipFile(file).read(Path(file).stem)), sheet_name=None)
        
        archive = ZipFile(file)

        # find file names in the archive which end in `.xls`, `.xlsx`, `.xlsb`, ...
        files_in_archive = archive.namelist()
        excel_files_in_archive = [
            f for f in files_in_archive if Path(f).suffix[:4] == ".xls"
        ]
        # ensure we only have one file (otherwise, loop or choose one somehow)
        assert len(excel_files_in_archive) == 1

        # read in data
        data_mapping = pd.read_excel(
            BytesIO(archive.read(excel_files_in_archive[0])),
            sheet_name=None, header=None,
        )

        
        
               row_counts = []
    for sheet in list(data_mapping.keys()):
        if sheet == 'Sheet1':
            df = data_mapping.get(sheet)[3:]
         
        else:
              df = data_mapping.get(sheet)
        row_counts.append(len(df))
        print(len(data_mapping.get(sheet)))


      
        
        

        file_mapping.update({file: sum(row_counts)})

    frame = pd.DataFrame([file_mapping]).transpose().reset_index()
    frame.columns = ["file_name", "row_counts"]

    return frame



dir_path = r'D:\test22 - 10'





zip_files = []
for root, dirs, files in os.walk(dir_path):
    for file in files:
        if file.endswith('.zip'):
            zip_files.append(os.path.join(root, file))
df = process_files(zip_files)   #function

does anyone have an idea on what im doing wrong?

You just need to use the skiprows argument:
https://pandas.pydata.org/docs/reference/api/pandas.read_excel.html

# read in data
data_mapping = pd.read_excel(
     BytesIO(archive.read(excel_files_in_archive[0])),
     sheet_name=None, header=None, skiprows=2
)

or don't use skiprows and then slice the sheet's dataframe directly:

row_counts = []
for sheet in list(data_mapping.keys()):
     if sheet == 'name of first sheet':
          df = data_mapping.get(sheet)[3:]
     else:
          df = data_mapping.get(sheet)
     row_counts.append(len(df))
     print(len(data_mapping.get(sheet)))

##or based on the location in the list. you don't need to call list() on .keys()
for sheet, i in enumerate(data_mapping.keys()):
     if i == 0:
          df = data_mapping.get(sheet)[3:]
     else:
          df = data_mapping.get(sheet)
     row_counts.append(len(df))
     print(len(data_mapping.get(sheet)))