laksh9950 Goto Github PK

followers: 36.0 following: 169.0 repos: 226.0 gists: 4.0

Name: Laxya Agarwal

Type: User

Company: IIT Bombay

Bio: Machine Learning Engineer | Pytorch | | Learning and Predicting Stuff

Location: Bangalore

Blog: laksh9950.github.io

I'm Laxya, a full-time full-stack Developer 👨‍💻 working remotely since 2015 🚀 . I help Startups deploy machine learning (Text , Vision and audio ) and blockchain solutions.

🌱 I’m currently learning how to optimize large language models and to scale billion parameter machine learning models on CPUs.
My latest project is based on real time utility of LLM (LLama 2), TTS( Text to Speech), STT (Speech to Text), and Text to Image .
Optimizing the models and delivering real time results have huge potential for end user applications.
❓ Ask me about anything related to Python , LLM ,Transformers , Text to specch , Diffusion models , LORA QLORA , Flash attention.
⚡ Fun fact: I deploy on friday

My Skill Set

Frontend

Backend

DevOps

Connect with me

Github Stats

Laxya Agarwal's Projects

stanford-cs-229-machine-learning

VIP cheatsheets for Stanford's CS 229 Machine Learning

stanford-cs229

🤖 Exercise answers to the problem sets from the 2017 machine learning course cs229 by Andrew Ng at Stanford

stanford-cs231n-2018

Solutions to Stanford CS231n Spring 2018 Course Assignments.

stanford_cs231n_2019

Solutions and comments to assignments for 2019 Stanford's course on convolutional neural networks

statrethinking_winter2019

Statistical Rethinking course at MPI-EVA from Dec 2018 through Feb 2019

A Neural Language Style Transfer framework to transfer natural language text smoothly between fine-grained language styles like formal/casual, active/passive, and many more. Created by Prithiviraj Damodaran. Open to pull requests and other forms of collaboration.

stylegan2encoder

StyleGAN2 - Official TensorFlow Implementation

super-mario-bros-a3c-pytorch

supply-chain-simulation-environment

system-design-101

Explain complex systems using visuals and simple terms. Help you prepare for system design interviews.

system-design-primer

Learn how to design large-scale systems. Prep for the system design interview. Includes Anki flashcards.

ta-style-introduction-to-web-and-html-css-thaaaa

talk

Let's make sand talk

technicaltuesdays

Repository for the slides of the Technical Tuesdays Seminar Series

tensorflow-serving_sidecar

Serve machine learning models using tensorflow serving

tensorflow2-generative-models

Implementations of a number of generative models in Tensorflow 2. GAN, VAE, Seq2Seq, VAEGAN, GAIA, Spectrogram Inversion. Everything is self contained in a jupyter notebook for easy export to colab.

tf2_course

Notebooks for my "Deep Learning with TensorFlow 2 and Keras" course

tfjs-models

Pretrained models for TensorFlow.js

tfrecord

TFRecord reader for PyTorch

the-book-of-secret-knowledge

A collection of inspiring lists, manuals, cheatsheets, blogs, hacks, one-liners, cli/web tools and more.

tinder-swipe-bot

tldraw

a very good whiteboard

tner

Language model finetuning on NER with an easy interface, and cross-domain evaluation. We released NER models finetuned on various domain via huggingface model hub.

trades

Track to Detect and Segment: An Online Multi-Object Tracker (CVPR 2021)

training-data-analyst

Labs and demos for courses for GCP Training (http://cloud.google.com/training).

transcribe

Transcribe is OpenAI's chatGPT based live transcription, conversation platform. It provides live transcripts from microphone input and audio output speaker. It optionally generates a suggested conversation response using OpenAI's GPT API. It is very effective in providing relevant responses for the current conversation.

transformers

🤗 Transformers: State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch.