Explore topics with us

What is RAG ?

Large language models have been long challenged by poor contextualisation, lower adaptability to specific cases, reliability and scalability issues. This is where fine tuning and RAG come into play.

RAG or Retrieval Augmented Generation is a technique being used by large language models to improve quality and accuracy of their output by providing a bigger reference or knowledge base to the model for responding to user's query.

The increased contextual awareness, scalability, reliability and lower computational costs are the major benefits the technique delivers.

How to treat missing values in a dataset?

Missing data points in a dataset is a common occurrence in real-world problems. Handling missing values is crucial to maintain the integrity and accuracy of data. Before we come to that, it is imperative to understand where is it coming from, is it missing randomly, and how does these impact the way one should address them.

Some practical tips for survey design

Survey design is very crucial to meet the objective of a study. Whereas it is important to have a good response rate for analysis, equally important is to gather accurate and meaningful data. Avoiding common pitfalls while designing a questionnaire can enhance data quality as well as the response rate.