Wsrtjtyk

I am a Python beginner and I encounter with a NLP project.

This is my code:

doc1 = "I am a dog and I like biscut"

doc2 = "I am a cat"

doc3 = "I like to drink milk"

doc4 = "I am a bird and I fly to the sky"

doc5 = "I am an elephant and I want to sleep"

docs = [doc1.split(' '), doc2.split(' '), doc3.split(' ')]

docs2 = [doc4.split(' '), doc5.split(' ')]

docs_all = doc1.split(' ') + doc2.split(' ') + doc3.split(' ') + 

doc4.split(' ') + doc5.split(' ')

And get a set of docs_all:

print(list(enumerate(set(docs_all))))

[(0, 'a'), (1, 'the'), (2, 'drink'), (3, 'elephant'), (4, 'dog'), (5, 

'biscut'), (6, 'cat'), (7, 'bird'), (8, 'an'), (9, 'milk'), (10, 'want'), 

(11, 'am'), (12, 'I'), (13, 'and'), (14, 'to'), (15, 'sky'), (16, 'sleep'), 

(17, 'like'), (18, 'fly')]

The reference matrix of docs and docs 2 is :

setdocs = [(0, 0.44), (1, 0.14), (2, 0.22), (3, 0.113), (4, 0.44), (5, 

0.15), (6, 0.96), (7, 0.77), (8, 0.28), (9, 0.39), (10, 0.111)]

setdocs2 = [(0, 0.55), (1, 0.13), (2, 0.52), (3, 0.33), (4, 0.114), 

(5,0.995),(6, 0.16), (7, 0.97), (8, 0.118), (9, 0.14), (10, 0.88), (11, 

0.166), (12, 0.85)]

The first value in the tuple is the number of words of docs and docs2, come from

refdocs2 = list(enumerate(set(doc4.split(' ') + doc5.split(' '))))

refdocs = list(enumerate(set(doc1.split(' ') + doc2.split(' ') + 

doc3.split(' '))))

print(refdocs)

print(refdocs2)

[(0, 'a'), (1, 'drink'), (2, 'dog'), (3, 'biscut'), (4, 'cat'), (5, 'milk'), 

(6, 'am'), (7, 'I'), (8, 'and'), (9, 'to'), (10, 'like')]

[(0, 'a'), (1, 'the'), (2, 'elephant'), (3, 'bird'), (4, 'an'), (5, 'want'), 

(6, 'sleep'), (7, 'am'), (8, 'I'), (9, 'and'), (10, 'to'), (11, 'sky'), (12, 

'fly')]

I want to get a matrix that is about:

finaldocs = [[0.44, 0, 0, 0, 0.22, 0.113, 0, 0, 0, 0, 0, 0.96, 0.77, 0.28, 

            0, 0, 0, 0.111, 0],

            [0.44, 0, 0, 0, 0, 0, 0.44, 0, 0, 0, 0, 0.96, 0.77, 0, 0, 0, 0, 

            0, 0],

            [0, 0, 0.14, 0, 0, 0, 0, 0, 0, 0.15, 0, 0, 0.77, 0, 0.39, 

            0, 0, 0.111, 0]]

finaldocs2 = [[0.55, 0.13, 0, 0, 0, 0, 0, 0.33, 0, 0, 0.97, 0.118, 0.14, 

             0.88, 0.166, 0, 0, 0.85],[0, 0.13, 0, 0.52, 0, 0, 0, 0, 0.114, 

             0.995, 0.97, 0.118, 0.14, 0.88, 0, 0.16, 0, 0]]

The second value in the tuple of setdocs and setdocs2 is the value I want to take out.
finaldocs[0] to finaldocs[3] refer to doc1 to doc3 and to get the second value of the tuple in setdocs with the ordinal number of list(enumerate(set(docs_all)))

For example, doc2 = "I am a cat" occur in the 0, 6, 11 ,12 value in list(enumerate(set(docs_all))). "I" "am" "a" "cat" occur in the 0,4,6,7 value in refdocs and get the second value in the tuple of from setdocs to create finaldocs[2]

[0.44, 0, 0, 0, 0, 0, 0.44, 0, 0, 0, 0, 0.96, 0.77, 0, 0, 0, 0, 0, 0]

My attempts:

dll = [np.arange(19),np.arange(19),np.arange(19)]

for i in dll:

for ii in i:

    for m in list(enumerate(set(docs_all))):

        for mm,nn in m:

            for t in refdocs:

                for tt,ll in t:

                    for p in setdocs:

                        for pp,oo in p:

                            if nn in ll:

                                i.replace(i, oo)

It does fail.

How do I get finaldocs by coding in Python?