Vibhav Inna Kedege

Vibhav Inna Kedege

Software Designer | Data Analyst | Runner | Photographer | Dog Lover

LLM Application - Q/A Assistant

less than 1 minute read

Published:

In this blogpost, I created an interactive command line Q&A assistant application that uses an Ollama LLM, vector database to answer user questions through a retriever augmented generation (RAG) approach

Medium Blog Link

You May Also Enjoy

Deep Learning in Tensorflow - The Perceptron

less than 1 minute read

Published:

In this blogpost, the fundamental building block (the perceptron) has been discussed to

  1. Translate the perceptron model to code
  2. Visualise the number of trainable hyper-parameters when number of layers and input changes

Basic Reinforcement Learning Algorithms

less than 1 minute read

Published:

In this blogpost, SARSA and Q-Learning Reinforcement Learning Algorithms have been used to solve classic OpenAI Gym problems of -

  1. Balancing a pole on a moving cart
  2. Driving a car up a mountain.

Autonomous Driving using Computer Vision

less than 1 minute read

Published:

In this blogpost, Convolution Neural Network based Computer Vision algorithms were used to train an autonomous virtual car.