How do I validate attribute values based on given map?-CodePudding

So, I have a set of pairs, let's call it map X, that looks this:

{'a':(a_1, a_2, a_3),
'b':(b_1, b_2, b_3),
'c':(c_1, c_2, c_3)}

and a table that looks something like this:

Attribute 1	Attribute 2
a	a_1
a	a_2
c	c_3
b	a_1
a	a_2
b	b_3

We can see that the fourth record associates b and a_1. However, a_1 does not belong to the list of values associated with b in map X. Generally speaking, I want to flag when something like this happens.

Using python (pandas, preferably, but plain python is okay), how do I confirm that each attribute 1 value is paired with a member of the collection of values associated with it in map X?

CodePudding user response：

First flatten dictionary, so possible mapping second column by Series.map and compare by first column if not equal:

d = {x: k for k, v in X.items() for x in v}

df['test'] = df['Attribute 2'].map(d).ne(df['Attribute 1'])
print (df)
  Attribute 1 Attribute 2   test
0           a         a_1  False
1           a         a_2  False
2           c         c_3  False
3           b         a_1   True
4           a         a_2  False
5           b         b_3  False

CodePudding user response：

You can try apply on rows

df['isin'] = df.apply(lambda row: row['Attribute 2'] not in d.get(row['Attribute 1']), axis=1)

print(df)

  Attribute 1 Attribute 2   isin
0           a         a_1  False
1           a         a_2  False
2           c         c_3  False
3           b         a_1   True
4           a         a_2  False
5           b         b_3  False