Adnan's Notes
Log Book
-
Audio Transformers - II
Discussion on Mel Spectogram.
-
Audio Transformer - I
Part one is dedicated to a discussion around the Apple Neural Engine. We will take a deep dive into the first principle discussed in [1]. In Part 2, we will examine the Self-Supervised Chewing Spectogram Transformer (SSCST), the audio preprocessing steps, deployment and quanitzation considerations. The architecture and use-case is for research purposes and should not be used for commercial applications without prior permission.
-
TLS
Transport Layer Security (TLS), is a protocol to encrypt data sent over the internet. If you imagine packets being sent over the internet as stacks take a look at the below picture to capture where TLS fits into that.
-
UCX and RDMA
UXC stands for Unified Communication X. It is a communication framework for high bandwidth, low latency networks. It exposes a set of abstract communication elements and is able to offload using RDMA, TCP, GPUs, shared memory etc.
-
HPC Systems and Docker
First, edit the settings.json file for VSCode. For Linux, do the following:
subscribe via RSS