Tip¶
rows with same values¶
# b is the index for rows with the same value in cols j:k
d = df.iloc[:,j:k].values
v = df.iloc[:,[j]].values
b = np.bitwise_and.reduce(d==v,axis=1)
np.where converted datetime64[ns] to object¶
avoid use np.where here; use pandas method (cap min to zero)