The following are the steps for executing TF-IDF in Python:

1. Import the library, as follows:

from sklearn.feature_extraction.text import TfidfVectorizer

2. Let's make a corpus by adding four documents, as follows:

corpus = ['First document', 'Second document','Third document','First and second document' ]

3. Let's set up the vectorizer:

vectorizer = TfidfVectorizer()

4. We extract the features out of the text as follows:

X = vectorizer.fit_transform(corpus)
print(vectorizer.get_feature_names())
print(X.shape)

The output is as follows:

5. Here comes the document term matrix; every list indicates a document:

X.toarray()

We get the following output: