Pytorch

A deep dive into how to use the Gym library to customize the agent’s environment and a PyTorch implementation of the Deep Q-Learning algorithm — I recently have started to play Qwixx with my girlfriend. It is a game with two dice and several colored ones. …

How to use 16bit float, TensorRT, network rewriting and multi-threading to dramatically speed up deep learning model prediction — Vision transformer such as UNET, SwinUNETR are state-of-the-art in computer vision tasks, such as semantic segmentation. But it takes a lot of time for such models to make a prediction. This article shows how to speed up such model’s prediction by 9 times. This improvement paves the way for many…

Deep learning has become a fundamental part of modern machine learning, and choosing the right library is crucial for success. JAX and PyTorch are two popular options, each with its unique features and advantages. In this article, we will dive deep into comparing JAX and PyTorch, exploring their ecosystems, programming…

Part 2 Modular Codes — Note: Before proceeding with this article, it is essential to complete Part 1, which provides a step-by-step guide on creating a neural network. In this article, we will revisit our Food Classification Project, this time implementing it in a modular style. …

Intention The intention of this article is NOT to walk through in steps the steps manner, but focus on the gotchas I faced and an unexpected performance result. Setup I have a conda environment that have most required package installed, and the most important one is upgrade to PyTorch 2.0.1 (I tried…

To convert a PyTorch model to TensorFlow Lite format, you’ll need to perform a few steps. Here’s an outline of the process: 1. Convert the PyTorch model to ONNX format: First, you’ll convert the PyTorch model to the ONNX (Open Neural Network Exchange) format. ONNX is an open standard for…

This is a workbook for the one who wants a quick start into PyTorch. What will be covered in this section? # Attributes `torch.ndim`, `tensor.shape`, `tensor.dtype`, `tensor.device` # Methods `torch.tensor()`, `scalar.item()`, `torch.rand()`, `torch.ones()`, `torch.zeros()`, `torch.zeros_like()`, `torch.ones()`, `torch.arange()`, `torch.ones_like()`, `tensor.type()` # Import PyTorch import torch # Scalar, Vector, Matrix, Tensor Creation. You will…

Introduction: Artificial Intelligence (AI) has become a buzzword in our digital era, shaping our lives and industries in unprecedented ways. It’s an exciting field that holds immense potential for innovation and progress. In this article, we will delve into some of the most famous AI tools and explore essential terms to…

A simple answer to “I can’t increase batch size because of limited GPU” — Many times while training deep neural networks, we wish we had a bigger GPU to accommodate the BIG Batch size .. but we don’t !! and it leads to: slower convergence and lesser accuracy Let’s revisit the typical training loop in PyTorch. …