gabecano4308’s gists

gabecano4308 / bball ref -- scatter

Created January 7, 2021 05:19

for part 2

	sns.scatterplot(all_teams_2021['Min PG'], all_teams_2021['Turnovers PG'])
	plt.title('Relationship Between Average Minutes Played and Turnovers Per Game');

gabecano4308 / bball ref -- hist

Created January 7, 2021 05:13

for part 2

	sns.histplot(all_teams_2021['Min PG'], bins=20)
	plt.title('Frequency of Minutes Played Per Game in 2021 Across the NBA');

gabecano4308 / bball ref -- groupby

Created January 7, 2021 05:07

for part 2

all_teams_2021.groupby('Team')[['Weight (Lbs)']].sum().sort_values(by='Weight (Lbs)', ascending=True).head(10)

gabecano4308 / bball ref -- det fg%

Created January 7, 2021 05:05

for part 2

all_teams_2021[(all_teams_2021['Team'] == 'DET') & (all_teams_2021['Field Goal %'] <= 0.40)]

gabecano4308 / bball ref -- obj to float

Last active January 7, 2021 04:17

for part 2

	# Function changes all object columns that should be numeric to an int or float
	def obj_to_num(nba_df):

	str_cols = ['Name', 'Team', 'Twitter Handle', 'Position', 'Height']

	for i in nba_df.columns:
	if i not in str_cols:
	nba_df[i] = pd.to_numeric(nba_df[i])

	print(nba_df.dtypes)

gabecano4308 / bball ref -- get season stats (all teams)

Created January 7, 2021 03:41

for part 2

	def get_stats(year):
	# Creating a list of dictionaries to then convert into a Pandas Dataframe
	nba_info = []

	# Iteratively finding the URL page for each NBA team according to the 'year' parameter and instantiating
	# a BeautifulSoup object
	for i in team_list:
	team_url = (f'https://www.basketball-reference.com{i}/{str(year)}.html')
	team_res = requests.get(team_url)
	team_soup = BeautifulSoup(team_res.content, 'lxml')

gabecano4308 / bball ref -- beautiful soup

Created December 30, 2020 20:11

for part 1

	# BeautifulSoup library parses the content of an HTML document, in this case wiz_res
	wiz_soup = BeautifulSoup(wiz_res.content, 'lxml')

	# BeautifulSoup's .find() method searches for a tag and specified attributes,
	# returning the first match
	wiz_per_game = wiz_soup.find(name = 'table', attrs = {'id' : 'per_game'})

gabecano4308 / bball ref -- combined df

Last active December 31, 2020 19:09

for part 1

	# URL for the Washington Wizards Basketball Reference page
	wiz_url = (f'https://www.basketball-reference.com/teams/WAS/2021.html')

	# The requests library can send a GET request to the wiz_url
	wiz_res = requests.get(wiz_url)

	# BeautifulSoup library parses the content of an HTML document, in this case wiz_res
	wiz_soup = BeautifulSoup(wiz_res.content, 'lxml')

	# BeautifulSoup's .find() method searches for a tag and specified attributes,

gabecano4308 / bball ref -- ht_wt_pos

Created December 29, 2020 23:32

for part 1

	height_weight_position = []

	for row in wiz_per_game.find_all('tr')[1:]:

	player = {}

	# Parsing html data from each player's specific web page
	player_url = ('https://www.basketball-reference.com/' + row.find('a').attrs['href'])
	player_rest = requests.get(player_url)
	player_soup = BeautifulSoup(player_rest.content, 'lxml')

gabecano4308 / bball ref -- scrape twitter

Last active December 31, 2020 18:11

for part 1

	twitter_handle = []

	for row in wiz_per_game.find_all('tr')[1:]:
	player = {}

	# Taking the row's first hyperlink (player's url ending) and appending it to the base url
	# to get the player's personal webpage url.
	player_url = ('https://www.basketball-reference.com/' + row.find('a').attrs['href'])

	# Making a new BeautifulSoup instance of the player's webpage and narrowing it to the top section