Pandas SQL equivalent for 'not equal' clause

Question

I do not see this in the SQL comparison documentation for Pandas. What would be the equivalent of this SQL in Pandas?

select a.var1, a.var2, b.var1, b.var2
from tablea a, tableb b
where a.var1=b.var1 
and a.var2=b.var2
and a.var3 <> b.var3

I have the merge code as follows:

df = pd.merge(a, b, on=['VAR1','VAR2'], how='inner')

How do I incorporate the 'not equal' portion?

and a.var3 <> b.var3

Answer 1

You can query the resulting frame:

a.merge(b, on=['VAR1','VAR2']).query('VAR3_x != VAR3_y')