data-preprocessors 0.26.0

An easy to use tool for Data Preprocessing specially for Text Preprocessing

Data Preprocessors

_{An easy to use tool for Data Preprocessing specially for Text Preprocessing}

Table of Contents

• Installation
• Quick Start
• Features
  • Split Textfile
  • Build Parallel Corpus
  • Separate Parallel Corpus
  • Remove Punctuation
  • Space Punctuation
  • Text File to List
  • List to Text File
  • Count Characters of a Sentence
  • Count Words of Sentence
  • Count No of Lines in a Text File
  • Apply Any Function in a Full Text File

Installation

Install the latest stable release
For windows

pip install -U data-preprocessors

For Linux/WSL2

pip3 install -U data-preprocessors

Quick Start

from data_preprocessors import text_preprocessor as tp
sentence = "bla! bla- ?bla ?bla."
sentence = tp.remove_punc(sentence)
print(sentence)

>> bla bla bla bla

Features

Split Textfile

This function will split your textfile into train, test and validate. Three separate text files. By changing shuffle and seed value, you can randomly shuffle the lines of your text files.

from data_preprocessors import text_preprocessor as tp
tp.split_textfile(
    main_file_path="example.txt",
    train_file_path="splitted/train.txt",
    val_file_path="splitted/val.txt",
    test_file_path="splitted/test.txt",
    train_size=0.6,
    val_size=0.2,
    test_size=0.2,
    shuffle=True,
    seed=42
)

# Total lines:  500
# Train set size:  300
# Validation set size:  100
# Test set size:  100

Separate Parallel Corpus

By using this function, you will be able to easily separate src_tgt_file into separated src_file and tgt_file.

from data_preprocessors import text_preprocessor as tp
tp.separate_parallel_corpus(src_tgt_file="", separator="|||", src_file="", tgt_file="")

Remove Punctuation

By using this function, you will be able to remove the punction of a single line of a text file.

from data_preprocessors import text_preprocessor as tp
sentence = "bla! bla- ?bla ?bla."
sentence = tp.remove_punc(sentence)
print(sentence)

# bla bla bla bla

Space Punctuation

By using this function, you will be able to add one space to the both side of the punction so that it will easier to tokenize the sentence. This will apply on a single line of a text file. But if we want, we can use it in a full twxt file.

from data_preprocessors import text_preprocessor as tp
sentence = "bla! bla- ?bla ?bla."
sentence = tp.space_punc(sentence)
print(sentence)

# bla bla bla bla

Text File to List

Convert any text file into list.

 mylist= tp.text2list(myfile_path="myfile.txt")

List to Text File

Convert any list into a text file (filename.txt)

tp.list2text(mylist=mylist, myfile_path="myfile.txt")

Count Characters of a Sentence

tp.count_chars(myfile="file.txt")

Apply a function in whole text file

In the place of function_name you can use any function and that function will be applied in the full/whole text file.

from data_preprocessors import text_preprocessor as tp
tp.apply_whole(
    function_name, 
    myfile_path="myfile.txt", 
    modified_file_path="modified_file.txt"
)