子集pandas DataFrame，并分为3个DataFrame

Question

How can I subset the pandas DataFrame by the values in one column? For example, I want to separate the dataset below by the names of each Company.

So I want to split the keywords data frame into 3 different data frames. I tried to def a function that would split the dataset by the name value in the column and then ran a for loop on the column for the function. However, it doesn't seem to work. Anyone know how I can accomplish this?

keywords = {'Company':['amazon', 'amazon', 'amazon', 'target' 'target' 'target', 'walmart', 'walmart', 'walmart'], 
'keywords':['abc', 'def', 'ghi', 'jkl', 'mno', 'pqr', 'rst', 'uvw', 'xyz'], 
'type':['article', 'blog', 'news', 'article', 'blog', 'news', 'article', 'blog', 'news']}

def key(name):
    key = keywords.loc[name, :]
    return

for h in keywords['Company']:
    key(h)

Answer 1

the following assumes df is a dataframe loaded with your keywords data

amazon_df = df.query('Company == "amazon"')

this will return a new dataframe where the company column matches the string 'amazon'. To pass in a variable to df.query string you prepand an @ symbol to the variable name. pandas.DataFrame.query docs

for example:

def get_subset_df(df, company_name):
    return df.query('Company == @company_name')