Surya Kant Sahu's repositories
Time2Vec-PyTorch
Reproducing the paper: "Time2Vec: Learning a Vector Representation of Time" - https://arxiv.org/pdf/1907.05321.pdf
SmoothedGradientDescentAscent
An Implementation of Smoothed Gradient Descent-Ascent (NeurIPS 2020) and a variant of the algorithm.
AttentionEntropy
Applying Principle of Least Action to Transformers to make them data efficient
Drug-Protein-Interaction-ML
This Repo contains 1D Convolution Based Model for Drug-Protein Interaction Prediction problem.
LearningInitializations
An attempt at making a model learn how to initialize other models to reduce training time/error.
MyMusicTransformer
RLHF + MusicTransformer = Generate the music YOU love
WordGuesser
Guessing the word given an incomplete word and its description
AlignedMusicTransformer
Aligning Music Transformer to Human Preferences (my music preferences to be precise)
AutoAWQ
AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference.
Compact-Transformers
[Preprint] Escaping the Big Data Paradigm with Compact Transformers, 2021
dialogy
Language understanding toolkit for human dialogs.
google-research
Google Research
HacktoberFestContribute
Contribute codes on Data Structures and Algorithms. #HacktoberFest
InsanelyFastStringQuantization
A Python Library for Extracting Text features using Quantization/Hashing. Insanely. Fast.
JacobianDropout
This Repository contains code for an ongoing research on a novel Dropout-like layer.
mdistiller
A Knowledge Distillation Toolbox. The official implementation of https://arxiv.org/abs/2203.08679
NIID-Bench
Federated Learning on Non-IID Data Silos: An Experimental Study
ojus1.github.io
Nothing much to add here, just go to https://ojus1.github.io
peft
🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.
PoorMansJobScheduler
A simple Commandline Tool to pass Asynchronous jobs to a Local/Remote Workstation. Meant for ML Tasks.
qml-mooc
Lecture notebooks and coding assignments for the quantum machine learning MOOC created by Peter Wittek on EdX in the Spring 2019
RWKV-LM
RWKV-2 is a RNN with transformer-level performance. It can be directly trained like a GPT transformer (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.
TripletTorch
Triplet Loss Utility for Pytorch Library.
trlx
A repo for distributed training of language models with Reinforcement Learning via Human Feedback (RLHF)
vit-pytorch
Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch
vllm
A high-throughput and memory-efficient inference and serving engine for LLMs