Pibrary framework: A package of reusable code for ML projects
pip install pibrary
- File Class: Read and write files in csv, json, and pickle formats.
- String Class: String manipulation functions.
- LoguruPro Class: Loguru logger with additional features.
- Timeit Decorator: Decorator to measure the execution time of a function.
- Log Table Method: Print a table in the log.
from pibrary.file import File
from pibrary.loguru import logger
from pibrary.string import String
# File Class
dataframe = File(file_path).read().csv()
File(file_path).write(dataframe).csv()
json_data = File(file_path).read().json()
File(file_path).write(json_data).csv()
pickle_data = File(file_path).read().pickle()
File(file_path).write(pickle_data).csv()
# Logger
@logger.timeit
def some_function(...):
...
data = [
["Item 1", "Value 1", "Description 1", "Extra 1"],
["Item 2", "Value 2", "Description 2", "Extra 2"],
["Item 3", "Value 3", "Description 3", "Extra 3"],
["Item 4", "Value 4", "Description 4", "Extra 4"],
]
# Log the timing data as a table
logger.log_table(data)
# String Class
new_text = String(text).lower().remove_digits().remove_punctuation().strip()
The full documentation of Pibrary is available at https://pibrary.readthedocs.io/en/latest/.
Contributions are welcome! Please read CONTRIBUTING for details on how to contribute to this project.
This project is licensed under the terms of the MIT license.