From 02e612f774ecd83ad8724a811738eb64ae6f73f2 Mon Sep 17 00:00:00 2001
From: Veronika Kirenkina
 <112555585+veronica-kirenkina@users.noreply.github.com>
Date: Tue, 14 Nov 2023 04:03:22 +0300
Subject: [PATCH 1/2] Create practice_7.py

---
 practice_7.py | 91 +++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 91 insertions(+)
 create mode 100644 practice_7.py

diff --git a/practice_7.py b/practice_7.py
new file mode 100644
index 0000000..82088f4
--- /dev/null
+++ b/practice_7.py
@@ -0,0 +1,91 @@
+from typing import List
+from math import log
+from HW_3 import CountVectorizer
+
+class TfidfTransformer:
+    """
+        Класс TfidfTransformer используется для оценки важности слова в контексте документа,
+        являющегося частью коллекции документов или корпуса.
+
+        Methods
+        -------
+        tf_transform(count_matrix)
+            Выводит матрицу с подсчетом количества повторений слова к общему количеству слов в документе
+        idf_transform(count_matrix)
+            Выводит массив с подсчетом количества всех документов к документам, где встречается нужное слово
+        fit_transform(count_matrix)
+            Преобразуйте матрицу подсчета в нормализованное представление tf-idf
+        """
+
+    @staticmethod
+    def tf_transform(count_matrix: List[int]) -> List[List[int]]:
+        tf_matrix = []
+        for doc in count_matrix:
+            cnt_words = sum(doc)
+            tf_vector = [round(word / cnt_words, 3) for word in doc]
+            tf_matrix.append(tf_vector)
+        return tf_matrix
+
+    @staticmethod
+    def idf_transform(count_matrix: List[int]) -> List[List[int]]:
+        docs_len = len(count_matrix[0])
+        cnt_docs = len(count_matrix)
+        idf_matrix = []
+        for word in range(docs_len):
+            cnt_docs_with_word = 0
+            for doc in count_matrix:
+                if doc[word]:
+                    cnt_docs_with_word += 1
+            idf = round(log((cnt_docs + 1) / (cnt_docs_with_word + 1)) + 1, 3)
+            idf_matrix.append(idf)
+
+        return idf_matrix
+
+    @classmethod
+    def fit_transform(cls, count_matrix: List[int]) -> List[List[int]]:
+        tf_idf_matrix = []
+        for tf_vec in cls.tf_transform(count_matrix):
+            tf_idf_matrix.append([round(i * j, 3) for i, j in zip(cls.idf_transform(count_matrix), tf_vec)])
+        return tf_idf_matrix
+
+
+class TfidfVectorizer(CountVectorizer):
+    """
+        Класс TfidfTransformer используется для преобразования коллекции необработанных документов
+        в матрицу объектов TF-IDF. Наследуется от класса CountVectorizer.
+
+        Attributes
+        ----------
+        tfidf_transformer : TfidfTransformer
+            экземпляр класса TfidfTransformer, необходимы для составления композиции
+
+        Methods
+        -------
+        fit_transform(corpus)
+            Возвращает матрицу применение tf-idf на корпусе документов
+        """
+
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.tfidf_transformer = TfidfTransformer()
+
+    def fit_transform(self, corpus: List[str]) -> List[List[int]]:
+        matrix = super().fit_transform(corpus)
+        return self.tfidf_transformer.fit_transform(matrix)
+
+
+if __name__ == '__main__':
+    corpus = [
+        'Crock Pot Pasta Never boil pasta again',
+        'Pasta Pomodoro Fresh ingredients Parmesan to taste'
+    ]
+    transformer = TfidfTransformer()
+    tfidf_matrix = transformer.fit_transform(count_matrix)
+    print(transformer.tf_transform(count_matrix))
+    print(transformer.idf_transform(count_matrix))
+    print(tfidf_matrix)
+
+    vectorizer = TfidfVectorizer()
+    tfidf_matrix = vectorizer.fit_transform(corpus)
+    print(vectorizer.get_feature_names())
+    print(tfidf_matrix)

From 853451ca3b599c20405161de8982d3372c430a31 Mon Sep 17 00:00:00 2001
From: Veronika Kirenkina
 <112555585+veronica-kirenkina@users.noreply.github.com>
Date: Wed, 15 Nov 2023 14:12:44 +0300
Subject: [PATCH 2/2] Update practice_7.py

---
 practice_7.py | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/practice_7.py b/practice_7.py
index 82088f4..83e06a7 100644
--- a/practice_7.py
+++ b/practice_7.py
@@ -79,13 +79,18 @@ def fit_transform(self, corpus: List[str]) -> List[List[int]]:
         'Crock Pot Pasta Never boil pasta again',
         'Pasta Pomodoro Fresh ingredients Parmesan to taste'
     ]
+    vectorizer = CountVectorizer()
+    count_matrix = vectorizer.fit_transform(corpus)
+    print(vectorizer.get_feature_names())
+    print(count_matrix)
+
     transformer = TfidfTransformer()
     tfidf_matrix = transformer.fit_transform(count_matrix)
     print(transformer.tf_transform(count_matrix))
     print(transformer.idf_transform(count_matrix))
     print(tfidf_matrix)
 
-    vectorizer = TfidfVectorizer()
-    tfidf_matrix = vectorizer.fit_transform(corpus)
-    print(vectorizer.get_feature_names())
+    tf_vectorizer = TfidfVectorizer()
+    tfidf_matrix = tf_vectorizer.fit_transform(corpus)
+    print(tf_vectorizer.get_feature_names())
     print(tfidf_matrix)