The rapid exploration of outer space has faced a problem related to the insufficient efficiency of modern rocket engines. As a solution to this problem, the concept of a jet engine based on new operating principles is proposed, using a combination of known physical laws and having advantages over known types of jet engines.The article presents the results of testing three modifications of jet engines based on new operating principles and their analysis. The article discusses the physical principles of operation of a jet engine based on new principles, its advantages and problems arising during its creation.
Unlock the power of Transformer Neural Networks and learn how to build your own GPT-like model from scratch. In this in-depth guide, we will delve into the theory and provide a step-by-step code implementation to help you create your own miniGPT model. The final code is only 400 lines and works on both CPUs as well as on the GPUs. If you want to jump straight to the implementation here is the GitHub repo.
Transformers are revolutionizing the world of artificial intelligence. This simple, but very powerful neural network architecture, introduced in 2017, has quickly become the go-to choice for natural language processing, generative AI, and more. With the help of transformers, we've seen the creation of cutting-edge AI products like BERT, GPT-x, DALL-E, and AlphaFold, which are changing the way we interact with language and solve complex problems like protein folding. And the exciting possibilities don't stop there - transformers are also making waves in the field of computer vision with the advent of Vision Transformers.
How do people usually write a server if they don't really care about performance? A program starts, then starts accepting incoming connections from clients and starts a new thread for each client, which is engaged in servicing this client. If you use framework, like Spring or Flask or Poco there, then it does something like this inside itself - the only difference is the threads can be reused, that is, taken from a certain pool. It's all quite convenient, but not too effective (and Spring is bad). Most likely, your threads serving clients do not live long and most of the time they are waiting either to receive data from the client or to send it to the client - that is, they are waiting for some system calls to return. Creating an OS thread is quite an expensive operation, as is context switching between OS threads. If you want to be able to serve a lot of customers efficiently, you need to come up with something else. For example, callbacks, but they are pretty inconvenient (though there are different opinions on this).
Another option is to use non-blocking I/O in combination with some kind of implementation of user-space threads (fibers). In this article I will show you how to write all this with your own hands.
This is the complete crash course on the Elliptic Curves Cryptography used in Bitcoin and other blockchains. With code samples and a live demo from scratch.
