PyTorch

PyTorch can be used for two different things:

Replacing NumPy-like operations with GPU-accelerated operations.
Building deep neural networks.

PyTorch is its ease of use and simplicity. PyTorch uses dynamic computation, which allows greater flexibility in building complex architectures. PyTorch uses Python concepts, such as classes, structures, and conditional loops, allowing us to build AI algorithms in a pure object-oriented fashion. PyTorch was primarily built for research.

Data loading tutorials

Data loading (Auto Encoder with 2 images in output) tutorial
Data loading (Auto Encoder with dictionary sample in output) tutorial
Data loading and trasformation to TENSOR (Auto Encoder with 2 images in output) tutorial
Data loading and trasformation to TENSOR (Auto Encoder with dictionary sample in output) tutorial
Data loading, batching, and shuffling (Auto Encoder with 2 images in output) tutorial
Data loading, batching, and shuffling (Auto Encoder with dictionary sample in output) tutorial

REFERENCES:

0. PyTorch;