yashmitha

I code and design | currently exploring various domains MLOps, Neural nets and GOlang maybe?

[yashmitha @yashmitha_s]

So I started building a browser game inspired by Witch Hat Atelier where players draw spells and watch them come to life :D

Just finished most of frontend in svelte.

Next challenge: making a spell parser and sigil recognition

[yashmitha @yashmitha_s]

Started a new project, "Bitnote" It's a p2p file-sharing system designed for university course notes. so today I managed to get the central server and intermediate proxy service up and running. Here's the plan :D (still working on it)

[yashmitha @yashmitha_s]

Transformers revolutionized deep learning, powering modern LLMs like GPT-4, Gemini and much more. In upcoming posts, I will cover the workings of each layer of the transformer. :D
(6/6)

[yashmitha @yashmitha_s]

What are Transformers, and why did they replace RNNs, LSTMs and GRUs? A short thread ⬇️

[yashmitha @yashmitha_s]

Summary:
- RNNs → Process sequentially, slow & forgetful
- LSTMs/GRUs → Better memory but still sequential
- Transformers → Fully parallel, long-range memory, scalable
(5/6)

[yashmitha @yashmitha_s]

@Bhanu_here @maybeshalinii Alrightt thankssss

[yashmitha @yashmitha_s]

@khalidansariX3 Tbh reels has decreased my attention span to an extent that I struggle to watch any long videos on YouTube, I just can't :/

[yashmitha @yashmitha_s]

@dhrumitt_ Yess
my team struggled with the backend during a hackathon. We could figure out the frontend but the backend was a pain honestly😭

[yashmitha @yashmitha_s]

@yukteaX Yes sort of, I have a lot of hairfall since I moved Bangalore

[yashmitha @yashmitha_s]

@thepunitmkr 2 weeks

[yashmitha @yashmitha_s]

@__ShubhamGarg__ Congratulations 🥳🥳

[yashmitha @yashmitha_s]

@tausiqtweets I used https://t.co/uYOjSmKFhI for the contact form

[yashmitha @yashmitha_s]

Linear Regression and Optimization: Linear regression employs mathematical techniques like least squares and optimization algorithms to find the best-fit line and optimise model parameters.(7/7) #LinearRegression #Optimization

[yashmitha @yashmitha_s]

Mathematical concepts are the backbone of training a machine learning model. Let's dive into the key ones used in the process! #MathInML #MachineLearning (1/7)

[yashmitha @yashmitha_s]

Regularisation: Techniques like L1 and L2 regularisation help prevent overfitting in ML models. They introduce mathematical penalties to control complexity and improve generalisation. (6/7)