him-lychee 0.2.3

Lychee Language Core: A lightweight, high-performance library for slang translation and NLP text cleaning (pre-processing).

Lychee Language Core (him-lychee) Version 0.2.0 - Developed by Himpadma "Him"

Lychee is a lightweight, highly optimized Python package designed to quickly process user-generated text. It provides robust, single-pass slang replacement and a powerful suite of text cleaning tools necessary for Natural Language Processing (NLP) tasks like Sentiment Analysis.

Installation pip install him-lychee

Post-Installation Setup (Required for Full NLP Features) To use the advanced features (Stopwords, Stemming, Lemmatization, SpaCy), you must download the required models once:

python -m nltk.downloader stopwords punkt wordnet python -m textblob.download_corpora python -m spacy download en_core_web_sm

Lychee Core Usage (SlangDictionary Class) The core SlangDictionary class provides robust, optimized slang replacement.

Method

Description

Example Usage

replace_slang_in_text(text)

Crucial for Data Cleaning. Replaces all recognized slang terms in a single string with their full meanings. Highly optimized using a single regex pass.

slang_core.replace_slang_in_text(text)

get_meaning(slang_term)

Finds the meaning of a given slang term (case-insensitive).

slang_core.get_meaning('BRB')

reverse_lookup(meaning)

Finds all slang terms that map to a specific meaning.

slang_core.reverse_lookup('Laugh out loud')

Pandas Example (Recommended Usage) import pandas as pd import lychee

slang_core = lychee.SlangDictionary() df = pd.DataFrame({'review': ['OMG, that pic is GOAT!', 'IDK why BRB took so long.']})

Apply the function across the entire DataFrame column for high speed

df['cleaned_review'] = df['review'].apply(slang_core.replace_slang_in_text)

NLP Cleaning Pipeline (TextCleaner Class) The TextCleaner class provides functions to prepare text for machine learning models.

cleaner = lychee.TextCleaner() text = "The
GOAT said: https://example.com/ LOL! 😃"

Function

Description

Example Usage

remove_html_tags(text)

Strips HTML markup from the text.

cleaner.remove_html_tags(text)

remove_urls(text)

Removes all web URLs (http, https, www).

cleaner.remove_urls(text)

remove_punctuation(text)

Removes standard punctuation marks.

cleaner.remove_punctuation(text)

clean_emojis(text, mode='replace')

Replaces emojis with text codes (e.g., 😃 -> :smiling_face:), or removes them if mode='remove'.

cleaner.clean_emojis(text, 'replace')

remove_stopwords(text)

Removes common stop words (e.g., 'a', 'the', 'is').

cleaner.remove_stopwords(text)

spelling_correction(text)

Corrects common misspellings (using TextBlob, can be slow).

cleaner.spelling_correction(text)

stem_words(text)

Reduces words to their root form (e.g., 'running' -> 'run').

cleaner.stem_words(text)

lemmatize_text(text)

Reduces words to their dictionary form (e.g., 'better' -> 'good').

cleaner.lemmatize_text(text)

tokenize(text, library='nltk')

Splits text into word tokens using either NLTK or SpaCy.

cleaner.tokenize(text, 'spacy')