Pandas functions for a 2-D histogram of a dataframe: one column's values become the columns, and values become counts of given {row, column} pairs)

pandas.md

• Col1 is the field that will be the "rows" index
• Col2 is the column whose values will become the new columns
• Col3 is any other column (assuming that other columns are always filled; .count() will only count cells where Col3 has a value)

df \
.groupby(['Col1', 'Col2']) \
[['Col3']] \
.count() \
.reset_index() \
.pivot('Col1', 'Col2', 'Col3')