从bs4制作pandas DataFrame时如何跳过列？

Question

I am trying to scrape a table off of a website using Python and BeautifulSoup4. I then want to output the table, but I want to skip the first 5 columns of the table. Here is my code

def scrape_data():
    url1 = "https://basketball-reference.com/leagues/NBA_2020_advanced.html"
    html1 = urlopen(url1)
    soup1 = bs(html1, 'html.parser')
    soup1.findAll('tr', limit = 2)
    headers1 = [th.getText() for th in soup1.findAll('tr', limit = 2)[0].findAll('th')]
    headers1 = headers1[5:]
    rows1 = soup1.findAll('tr')[1:]
    player_stats = [[td.getText() for td in rows1[i].findAll('td')]for i in range(len(rows1))]
    stats1 = pd.DataFrame(player_stats, columns=headers1)
    return stats1

And the error I get is ValueError: 24 columns passed, passed data had 28 columns

I know the error is coming from stats1 = pd.DataFrame(player_stats, columns=headers1)

But how do I fix it?

Answer 1

Just use iloc on the resulting dataframe. Note that read_html returns a list of dataframes, although there is only one per this url. You need to access this single dataframe via pd.read_html(url)[0] . Then just use iloc to ignore the first five columns.

url = "https://basketball-reference.com/leagues/NBA_2020_advanced.html"
df = pd.read_html(url)[0].iloc[:, 5:]

Answer 2

I solved it thanks to some help from @JonClements. My working code is

def scrape_data():
    url1 = "https://basketball-reference.com/leagues/NBA_2020_advanced.html"
    html1 = urlopen(url1)
    soup1 = bs(html1, 'html.parser')
    soup1.findAll('tr', limit = 2)
    headers1 = [th.getText() for th in soup1.findAll('tr', limit = 2)[0].findAll('th')]
    headers1 = headers1[5:]
    rows1 = soup1.findAll('tr')[1:]
    player_stats = [[td.getText() for td in rows1[i].findAll('td')[4:]]for i in range(len(rows1))]
    stats1 = pd.DataFrame(player_stats, columns=headers1)
    return stats1