Skip to content

PyroNN v-1.1.0 documentation

PyPI version

Download PowerPoint Presentation

We will demonstrate the capabilities of Pyro-NN, a differentiable reconstruction framework.

  1. 📘 Introduction to Pyro-NN: Gain a basic understanding of the theory behind Pyro-NN.
  2. ⚙️ Installation and Setup: Learn how to install and set up Pyro-NN on your machine.
  3. 🔬 Working with Projection Data: Explore how to work with projection data to initiate a reconstruction.
  4. 🚀 Advanced Examples: Discover more advanced examples showcasing the usage of Pyro-NN.

1. 📘 Introduction to Pyro-NN

1.1 Motivation 💡

  • We can make use of known operators

Known Operators

  • For Deep Learning, the loss function and the amount of parameters to train can be reduced 🎯
  • We can have gradient flow through different domains 🔄
  • Parts of the neural network get interpretable, e.g. as filters 🧐

1.2 Basic Overview 🌐

Overview

General Notes 📋

- 🤖 Supports TensorFlow and PyTorch
- 💥 Full GPU Integration
- 🔓 Open Source
- 📜 Apache 2.0 License