DATA100-L3: Pandas Ⅰ

DataFrames: a data structure for tabular data

API

always remember to turn to GPT/google/doc

indexing and loc/iloc

generate subsets:

• loc: an operator select items by labels

  1	df.loc[row_indexer, column_indexer]

  • row_indexer: can be a single label, a list of labels, slice（闭区间）, single value
  • column_indexer: same as row_indexer
  • returns a DataFrame or Series

• iloc: an operator select items by positions

  1	df.iloc[row_indexer, column_indexer]

  • row_indexer: numeric index or a list of numeric indices，此时回到python经典索引 左闭右开
  • column_indexer: same as row_indexer
  • returns a DataFrame or Series 通常情况下，我们使用 loc 进行索引

• .head(6) and .tail() to get the first or last few rows of a DataFrame(syntactic sugar)

[ ]: context sensitive operator

series: a data structure for 1D labeled data

index: a array-like object that labels the rows and columns of a DataFrame

columns: usually do not have same name.

转换：

conditional selection

类型意识!

三种数据类型中的哪一个？？？

describe()

sample()

1

df.sample(n=5, replace=True) # randomly select 5 rows

value_counts()

1

df['column_name'].value_counts() # count the frequency of each value in a column

return a Series with the count of each value in the column.

unique()

1

df['column_name'].unique() # get all unique values in a column

return a numpy array with the unique values in the column.

sort_values()

1

df.sort_values(by='column_name', ascending=False) # sort the DataFrame by values in a column

return a new DataFrame with the rows sorted by values in a column.

reference

https://www.textbook.ds100.org/ch/a04/ref_pandas.html

https://pandas.pydata.org/pandas-docs/stable/reference/index.html