import unittest
import re
import nltk
from nltk import pos_tag
from nltk.tokenize import word_tokenize
from nltk.corpus import stopwords
from nltk.stem import WordNetLemmatizer
from nltk.corpus import wordnet as wn


# Helper functions
def _require_success(test_case, condition, failure_message):
    if not condition:
        test_case._testMethodName = failure_message
        test_case.fail()

def _check_success(test_case, condition, success_message, failure_message):
    if condition:
        test_case._testMethodName = success_message
    else:
        test_case._testMethodName = failure_message
        test_case.fail()

# Function that SHOULD NOT be modified
def get_wordnet_pos(treebank_tag):
    if treebank_tag.startswith('J'):
        return wn.ADJ
    elif treebank_tag.startswith('V'):
        return wn.VERB
    elif treebank_tag.startswith('R'):
        return wn.ADV
    else:
        return wn.NOUN

class TestUserCode(unittest.TestCase):
    # Test that everything is imported correctly
    def test_imports(self):
        import user_code
        _require_success(
            self, 
            hasattr(user_code, "word_tokenize"), 
            "`word_tokenize` is not declared"
        )
        _require_success(
            self,
            user_code.word_tokenize == word_tokenize,
            "`word_tokenize` is declared, but it's not `nltk.tokenize.word_tokenize`"
        )
        _require_success(
            self, 
            hasattr(user_code, "stopwords"), 
            "`stopwords` is not declared"
        )
        _require_success(
            self,
            user_code.stopwords == stopwords,
            "`stopwords` is declared, but it's not `nltk.corpus.stopwords`"
        )
        _require_success(
            self, 
            hasattr(user_code, "WordNetLemmatizer"), 
            "`WordNetLemmatizer` is not declared"
        )
        _require_success(
            self,
            user_code.WordNetLemmatizer == WordNetLemmatizer,
            "`WordNetLemmatizer` is declared, but it's not `nltk.stem.WordNetLemmatizer`"
        )
        _require_success(
            self, 
            hasattr(user_code, "pos_tag"), 
            "`pos_tag` is not declared"
        )
        _require_success(
            self,
            user_code.pos_tag == pos_tag,
            "`pos_tag` is declared, but it's not `nltk.pos_tag`"
        )
        _require_success(
            self, 
            hasattr(user_code, "wn"), 
            "`wn` is not declared"
        )
        _require_success(
            self,
            user_code.wn == wn,
            "`wn` is declared, but it's not `nltk.corpus.wordnet`"
        )
        _require_success(
            self,
            hasattr(user_code, "nltk"),
            "`nltk` is not declared"
        )
        _require_success(
            self,
            user_code.nltk == nltk,
            "`nltk` is declared, but it's not `nltk` library"
        )
        self._testMethodName = "imports are correct"

    # Test that required NLTK resources are downloaded
    def test_nltk_downloads(self):
        import user_code
        _require_success(
            self,
            hasattr(user_code, "nltk"),
            "`nltk` is not declared"
        )
        _require_success(
            self,
            user_code.nltk == nltk,
            "`nltk` is declared, but it's not `nltk` library"
        )
        with open("user_code.py", "r") as f:
            user_code_text = f.read()
            _require_success(
                self,
                re.search(r"nltk *\. *download *$ *'punkt_tab' *$", user_code_text),
                "Code for downloading `'punkt_tab'` shouldn't be modified"
            )
            _require_success(
                self,
                re.search(r"nltk *\. *download *$ *'stopwords' *$", user_code_text),
                "Code for downloading `'stopwords'` shouldn't be modified"
            )
            _require_success(
                self,
                re.search(r"nltk *\. *download *$ *'averaged_perceptron_tagger_eng' *$", user_code_text),
                "Code for downloading `'averaged_perceptron_tagger_eng'` shouldn't be modified"
            )
            _require_success(
                self,
                re.search(r"nltk *\. *download *$ *'wordnet' *$", user_code_text),
                "Code for downloading `'wordnet'` is not provided"
            )
        self._testMethodName = "NLTK resources are downloaded"

    # Test get_wordnet_pos
    def test_get_wordnet_pos(self):
        import user_code
        _require_success(
            self, 
            hasattr(user_code, "get_wordnet_pos"), 
            "`get_wordnet_pos` is not declared"
        )
        _require_success(
            self, 
            callable(user_code.get_wordnet_pos), 
            "`get_wordnet_pos` is not a function"
        )
        pos_tags = pos_tags = ['CC', 'CD', 'DT', 'EX', 'FW', 'IN', 'JJ', 'JJR', 'JJS', 'LS', 'MD', 'NN', 'NNS', 'NNP', 'NNPS', 'PDT', 'POS', 'PRP', 'PRP$', 'RB', 'RBR', 'RBS', 'RP', 'SYM', 'TO', 'UH', 'VB', 'VBD', 'VBG', 'VBN', 'VBP', 'VBZ', 'WDT', 'WP', 'WP$', 'WRB']
        _check_success(
            self, 
            list(map(user_code.get_wordnet_pos, pos_tags)) == list(map(get_wordnet_pos, pos_tags)),
            "`get_wordnet_pos` is correctly defined",
            "`get_wordnet_pos` shouldn't be modified"
        )
    
    # Test original text
    def test_text(self):
        import user_code
        expected = """Late in the evening, as the stars began to twinkle in the clear night sky, 
an old man sat quietly on the porch, reminiscing about his youthful adventures."""
        _require_success(
            self, 
            hasattr(user_code, "text"), 
            "`text` is not declared"
        )
        _check_success(
            self,
            user_code.text == expected,
            "`text` contains correct value",
            "`text` shouldn't be modified"
        )

    # Test lowercase conversion
    def test_text_lower(self):
        import user_code
        _require_success(
            self, 
            hasattr(user_code, "text"), 
            "`text` is not declared"
        )
        _require_success(
            self, 
            isinstance(user_code.text, str), 
            "`text` is not a string"
        )
        _require_success(
            self, 
            hasattr(user_code, "text_lower"), 
            "`text_lower` is not declared"
        )
        _check_success(
            self,
            user_code.text_lower == user_code.text.lower(),
            "`text_lower` is computed correctly",
            "`text_lower` is not the lowercase of `text`"
        )

    # Test tokenization of text_lower
    def test_tokens(self):
        import user_code
        _require_success(
            self, 
            hasattr(user_code, "text_lower"), 
            "`text_lower` is not declared"
        )
        _require_success(
            self, 
            isinstance(user_code.text_lower, str), 
            "`text_lower` is not a string"
        )
        _require_success(
            self, 
            hasattr(user_code, "tokens"), 
            "`tokens` is not declared"
        )
        _check_success(
            self,
            user_code.tokens == word_tokenize(user_code.text_lower),
            "`tokens` is computed correctly",
            "`text_lower` is not correctly tokenized into words"
        )

    # Test stop_words set
    def test_stop_words(self):
        import user_code
        _require_success(
            self, 
            hasattr(user_code, "stop_words"), 
            "`stop_words` is not declared"
        )
        _require_success(
            self, 
            isinstance(user_code.stop_words, set), 
            "`stop_words` is not a set"
        )
        _check_success(
            self,
            user_code.stop_words == set(stopwords.words('english')),
            "`stop_words` contains correct English stopwords set",
            "`stop_words` doesn't contain English stopwords"
        )

    # Test filtered_tokens filtering
    def test_filtered_tokens(self):
        import user_code
        _require_success(
            self, 
            hasattr(user_code, "tokens"), 
            "`tokens` is not declared"
        )
        _require_success(
            self, 
            isinstance(user_code.tokens, list), 
            "`tokens` is not a list"
        )
        _require_success(
            self, 
            hasattr(user_code, "stop_words"), 
            "`stop_words` is not declared"
        )
        _require_success(
            self, 
            isinstance(user_code.stop_words, set), 
            "`stop_words` is not a set"
        )
        _require_success(
            self, 
            hasattr(user_code, "filtered_tokens"), 
            "`filtered_tokens` is not declared"
        )
        expected = [tok for tok in user_code.tokens if tok not in user_code.stop_words]
        _check_success(
            self,
            user_code.filtered_tokens == expected,
            "`filtered_tokens` is computed correctly",
            "stopword in `tokens` are not correctly filtered out"
        )

    # Test tagged_tokens tagging
    def test_tagged_tokens(self):
        import user_code
        _require_success(
            self, 
            hasattr(user_code, "filtered_tokens"), 
            "`filtered_tokens` is not declared"
        )
        _require_success(
            self, 
            isinstance(user_code.filtered_tokens, list), 
            "`filtered_tokens` is not a list"
        )
        _require_success(
            self, 
            hasattr(user_code, "tagged_tokens"), 
            "`tagged_tokens` is not declared"
        )
        _check_success(
            self,
            user_code.tagged_tokens == pos_tag(user_code.filtered_tokens),
            "`tagged_tokens` is computed correctly",
            "`filtered_tokens` are not correctly pos tagged"
        )

    # Test lemmatizer instance
    def test_lemmatizer(self):
        import user_code
        _require_success(
            self, 
            hasattr(user_code, "lemmatizer"), 
            "`lemmatizer` is not declared"
        )
        _check_success(
            self,
            isinstance(user_code.lemmatizer, WordNetLemmatizer),
            "`lemmatizer` is defined correctly",
            "`lemmatizer` is not an instance of `WordNetLemmatizer`"
        )

    # Test lemmatizing results
    def test_lemmatized_tokens(self):
        import user_code
        _require_success(
            self, 
            hasattr(user_code, "lemmatizer"), 
            "`lemmatizer` is not declared"
        )
        _require_success(
            self,
            isinstance(user_code.lemmatizer, WordNetLemmatizer),
            "`lemmatizer` is not an instance of `WordNetLemmatizer`"
        )
        _require_success(
            self, 
            hasattr(user_code, "filtered_tokens"), 
            "`filtered_tokens` is not declared"
        )
        _require_success(
            self, 
            isinstance(user_code.filtered_tokens, list), 
            "`filtered_tokens` is not a list"
        )
        _require_success(
            self, 
            hasattr(user_code, "lemmatized_tokens"), 
            "`lemmatized_tokens` is not declared"
        )
        expected = [user_code.lemmatizer.lemmatize(tok, get_wordnet_pos(tag)) for tok, tag in user_code.tagged_tokens]
        _check_success(
            self,
            user_code.lemmatized_tokens == expected,
            "`lemmatized_tokens` is computed correctly",
            "`filtered_tokens` are not correctly lemmatized"
        )

    # Test print statement
    def test_print(self):
        with open("user_code.py", "r") as f:
            user_code_text = f.read()
        _check_success(
            self,
            re.search(r"print *$ *' ' *\. *join *\( *lemmatized_tokens *$", user_code_text),
            "print statement is correct",
            "print statement shouldn't be modified"
        )


if __name__ == '__main__':
    nltk.download('punkt_tab')
    nltk.download('stopwords')
    nltk.download('averaged_perceptron_tagger_eng')
    nltk.download('wordnet')
    unittest.main()

test_main.py

Explore os fundamentos do Processamento de Linguagem Natural (PLN) ao aprender técnicas essenciais de pré-processamento de texto e métodos para representação de dados textuais. Adquira experiência prática com ferramentas utilizadas para limpar, analisar e interpretar informações textuais. Desenvolva as habilidades necessárias para transformar linguagem bruta em insights estruturados, estabelecendo uma base sólida para aplicações avançadas em inteligência artificial e aprendizado de máquina.

Aprofunde-se nos fundamentos do pré-processamento de texto para preparar textos brutos para análise. Aprenda a tokenizar textos, filtrar stop words e personalizar a tokenização com expressões regulares.

Descubra como as palavras podem ser reduzidas às suas formas básicas utilizando stemming e lematização. Domine a marcação de classe gramatical para enriquecer o texto com contexto gramatical e aplique lematização sensível à classe gramatical.

Aprenda como textos podem ser representados por números utilizando modelos de espaço vetorial. Experimente na prática implementando e customizando dois modelos vetoriais populares: bag of words e TF-IDF.

Obtenha uma compreensão sólida sobre embeddings de palavras e como eles capturam significados semânticos. Explore as arquiteturas CBoW e Skip-gram utilizadas no Word2Vec e implemente-as por conta própria.

Desafio: Lematização com Marcação de POS

Solução

Desafio: Lematização com Marcação de POS

Solução