I want to read a HTML table as Pandas. But it is trickier than normal... Instead of the 2 headers (th) for some reason, the first column of each row is also indexed as th