alexanderholt

def gen_hist(df):  # plot all relationships
    heads = list(df.columns)
    while heads:
        base = heads.pop(0)
        for i in heads:
            sns.distplot(df[base])
            plt.show()
gen_hist(df_num)

alexanderholt

def eda(dataframe):
  # this code is from @ritikabhasker, slightly adapted, but mainly hers.
    print("**MISSING VALUES** \n", dataframe.isnull().sum(), "\n")
    print("**DATAFRAME INDEX** \n", dataframe.index, "\n")
    print("**DATAFRAME TYPES** \n", dataframe.dtypes, "\n")
    print("**DATAFRAME SHAPE** \n", dataframe.shape, "\n")
    print("**DATAFRAME DESCRIBE** \n", dataframe.describe(), "\n")
    print("**NUMBER OF UNIQUE VALUES PER COLUMN**")
    for item in dataframe:
        print(item)

alexanderholt

# Include this after you import matplotlib

import matplotlib.pyplot as plt
plt.rcParams['savefig.dpi'] = 500


.
.
.

alexanderholt

# Include this after you import matplotlib

import matplotlib.pyplot as plt
plt.rcParams['savefig.dpi'] = 500


.
.
.

alexanderholt

test_strip = 'Braund, Mr. Owen Harris'
test_strip.split('.')[1].lstrip()
# . split splits string into two tuples on either side of string
# then I pull the second tuple
#.lstrip removes leading white space by default


bad[i].split('(')[1].lstrip().rstrip(')')

alexanderholt

# import regex

#generate code from https://regex101.com

import re

regex = r"\d+"

for book in html.find_all('div', class_='booking'):
    matches = re.search(regex, book.text) #make sur eyou change from find iter

alexanderholt

#Joe says to try and except every time

import pandas as pd
dc_eats = pd.DataFrame(columns=["name","location","price","bookings"])

# loop through each entry
for entry in html.find_all('div', {'class':'result content-section-list-row cf with-times'}):
    # grab the name
    name = entry.find('span', {'class': 'rest-row-name-text'}).text
    # grab the location

alexanderholt

# clicking the 'next' button in selenium

link = driver.find_element_by_link_text('Next')
link.click()

alexanderholt

string = 'some string of characters'

from nltk.stem import WordNetLemmatizer
#instantiate
lemmatizer = WordNetLemmatizer()

#Before we can lemmitize our spam string we need to tokenize it.

from nltk.tokenize import RegexpTokenizer
tokenizer = RegexpTokenizer(r'\w+')

alexanderholt

string = 'string of charactrers

hvec = HashingVectorizer()
hvec.fit([spam])

df  = pd.DataFrame(hvec.transform([string]).todense())
df.transpose().sort_values(0, ascending=False).head(10).transpose()