- Hi, I’m Abdul. I am currently pursuing a Masters degree at University of Regensburg
- I publish blogs and my ideas on this website
- You can reach out to me at abdulrahimhere[at]yahoo.com
- Thank you for visiting and enjoy reading☕️
How Does Retrieval Augmented Generation Work
Introduction RAG (Retrieval-Augmented Generation) is a technique where a model can generate text based on external information retrieved from a knowledge base, rather than relying solely on its pre-trained knowledge. The knowledge base is a collection of documents $$ D = { d_1, d_2, \dots, d_n } $$ such that, whenever text is generated, the information is based on the knowledge base as much as possible. There are two components to a RAG model: ...
How to Build Neural Networks From Scratch
Part 1: Understanding the Pieces A convolutional neural network (CNN) works by scanning images for certain patterns like edges or textures. Each such scanner is called a kernel, which is nothing but a small matrix (usually 3x3 or 7x7) with floating point numbers (also called parameters). A kernel scans an image by sliding over it pixel by pixel. At each pixel, the weighted average of the neighbouring pixels is calculated using the weights from the kernel. This operation is called convolution. It produces another new image of often smaller dimentions. ...
Widecolumn database systems
Introduction Wide column database systems store data in column families. Below is a small example of a NoSQL database: Characterstics Although, these relations look like tables, but they’re not actually tables. This is because, we cannot make arbitrary selections based on non-primary columns. For example, the following query will fail: Therefore, similar to key-value stores, we can only query by using primary keys. A primary key in a wide column database consists of one or more partition keys and zero or more clustering keys (or sorting keys). ...
Can quantum entanglement help you sync your database replicas?
Introduction To ensure scalability, reliability and fault tolerance, web applications replicate data across multiple database nodes/clusters/availability zones/regions and cloud providers. However, replicating data introduces the challenge of keeping the copies in sync. In this article, we contemplate weather quantum entanglement can enable database replicas to sync instantinously. The ideas behind entanglement stem from quantum mechanics, a subfield of physics which describes the behaviour of sub atomic particles. In this post, we first describe the basics of database replication, then we move on the show the theoritical application of quantum entanglement in database replication. ...
Search completetion systems
Introduction When you type something on a search engine like google. You are continously shown search suggestions, as you type. API As you type in the search box, behind the scenes, google continously sends request to the backend. The requests might look like this: 1 2 3 4 5 GET /autocomplete?q=be { // metadata such as user id etc for personalized suggestion } The backend responds with a list of suggestions, which might look something like: ...