Deep learning library documentation
DLL is a deep learning library inspired by TensorFlow, PyTorch, and scikit-learn. It encompasses a wide range of deep learning and machine learning methods and includes numerous examples and tests to demonstrate their usage.
This library is intended as an educational project. While it offers a variety of functionalities, its performance and efficiency may not match those of the aforementioned libraries. Therefore, for production-level applications, it is recommended to use TensorFlow, PyTorch, or scikit-learn. However, DLL aims to provide greater clarity and ease of understanding compared to other libraries.
Library Structure
The library is divided into three main packages: Data, DeepLearning, and MachineLearning. Additionally, there is a fourth package for internal exceptions. Below is a brief overview of these packages:
Data: Contains utilities for data preprocessing, transformation, data loading, splitting, and assessing performance.
DeepLearning: Implements various deep learning architectures, which can be combined in various ways.
MachineLearning: Provides implementations of traditional machine learning algorithms. The package is divided into supervised and unsupervised learning.
Exceptions: Defines internal error handling mechanisms for the library.
For detailed documentation or some example scripts, refer to the sections below.