fgg blog

FAISS-IVFPQ

# Plain and Simple: IndexFlatL2

Given a set of vectors, we can index them using Faiss — then using another vector (the query vector), we search for the most similar vectors within the index. Now, Faiss not only allows us to build an index and search — but it also speeds up search times to ludicrous performance levels.

IndexFlatL2 measures the L2 (or Euclidean) distance between all given points between our query vector, and the vectors loaded into the index. It’s simple, very accurate, but not too fast.

IMG:indexFlat2

Image credit: pinecone.io

Noise_in_DSP

## What’s in a “Noise”?

  • As in English, it can mean an unwanted signal of any kind. If two signals interfere with each other, each signal would consider the other to be noise.
  • “Noise” also refers to a signal that contains components at many frequecies, so it lacks the harmonic structure of the periodic signals.

This post is about the second kind.

sampling and aliasing

模拟信号(analog signal)和数字信号(digital signal)是两种不同类型的信号,在电子通讯和 信号处理领域中有重要作用。

模拟信号是连续变化的信号,它的取值是在一定范围可以无限细分的(连续型变量)。例如声音波形。

数字信号是离散的信号,是对模拟信号进行离散化(采样和量化)得到的。进行这种离散化操作的叫 做模数转换器(ADC),转成数字信号后方便存储和传输,但利用这些数字信号时通常需要再将它们转 换为模拟信号(通过数模转换器,DAC),因为人类不像计算机,不能直接处理0/1信号,只能接收模 拟信号。(人的耳朵解析的是模拟信号,恐怕还没人练就听一串01组成的数字就能听出那是贝多芬的 第九交响曲的本领。)