obj : the object to be merged, e.g. [df1, df2]
axis : the direction of the merge
join : the way to merge, outer is an outer link, take the intersection
join_axes : set the column names to be displayedignore_index : whether to ignore the indexes of the original DataFrame/Series object and rearrange it
keys : set multiple keys for the data source. strong> : set multi-level index labels for the data source
levels :Specifies the indexes to be used as levels (inner indexes) of the hierarchical index, if keys is set
names :Names used to create hierarchical levels, if keys or levels is set
names :Names used to create hierarchical levels, if keys or levels is set. keys or levels
verify_integrity : Checks for duplicate indexes and raises an exception
left : the left DataFrame involved in the merge
right : the right DataFrame involved in the merge
how : joins: 'inner' (default, intersection); also, 'outer ', 'left', 'right'
on : the name of the column used for the join, which must be present in both left and right DataFrame objects, if specified, the intersection of the left and right column names is used as the join key
left_on : the column used as the join key in the left DataFarme
right_on : the column used as the join key in the right Column used as join key in DataFarme
left_index : Use the left row index as its join key
right_index : Use the right row index as its join key
sort : Sort the merged data by concatenation key, defaults to True, sometimes disable for better performance when working with large datasets
suffixes : A tuple of string values to append to the end of the overlapping columns, defaults to (''
''). The default is ('_x','_y'). For example, if both the left and right DataFrame objects have 'data', the result will be 'data_x', 'data_y'
copy : Set to False to avoid copying data into the result data structure in some special cases. By default, it is always assigned
It's easy to get started, and previous bloggers have written about it in great detail
Pandas Explained XV - Grouping with GroupBy Technology
A few additional points:
data : A column of the DataFrame used to create the pivot table. data, just enter the name of the column
index : row grouping tag
columns : column grouping tag
aggfunc : aggregate calculation method, defaults to (mean). Dictionary can be used to specify different aggregation functions for different columns, where data can be missing
fill_value : fill missing values
dropna : drop missing values
margins : whether or not the margins are removed
margins : whether or not the margins are removed
margins : whether or not the margins are removed. strong> : whether or not to aggfunc summarize the margins
margins_name : the name of the margin row/column