Reference image and article address: [The next big thing: the use of graph neural networks to discover particles](https://news.fnal.gov/2020/09/the-next-big-thing- the-use-of-graph-neural-networks-to-discover-particles/)

References What are lung opacities?

In the last week, I have been optimizing the Covid-Net I developed with my thesis and other research materials to diagnose the deep learning model of Covid19, and the accuracy I got from training on 20,000+ xRay datasets is less than 90%. When the model became more and more complex, the video memory was obviously not enough, so I sold the DJI Inspire 2 and used the money I earned to buy a TITAN RTX 24G, and then continued to optimize the model and increase training. of the batch_size.

The animation program code is as follows:

I have written an article about how to build a distributed training cluster on Raspberry Pie 4 using the distributed training system that comes with TF2.0. However, there is a drawback: we need to start the training program at each node, and the distributed training will only work after all the nodes are started. MPI is mainly used in the field of supercomputing. Building MPI cluster on Raspberry, firstly, it can be used to learn distributed computing on supercomputing, and secondly, it can…

Tensorflow Converter can convert models to FlatBuffers format. FlatBuffers is a cross-platform serialization tool where structured data is stored in binary form and can be used in the following ways The performance on the microcontroller is memory saving.

本文用到的Mathematica notebook源码可以在这里下载