Ayush Thakur's repositories
neurips-llm-efficiency-challenge
Starter pack for NeurIPS LLM Efficiency Challenge 2023.
kagglerecipes
Helpful data preprocessing, training, and visualisation code and scripts for a range of Kaggle competitions, supported by Weights & Biases.
llm-eval-sweep
A simple repository showcasing a few LLM Evaluation strategies and leverages W&B Sweeps to optimize the LLM system.
T5-RLHF-TF
Implementation of Reinforcement Learning from Human Feedback for Summarization Task in TensorFlow
kaggle-pogchamp
Classification pipeline for Kaggle Pogchamp Series competition built using TF/Keras and W&B.
mmdetection
OpenMMLab Detection Toolbox and Benchmark
HF-RL-Course
My exploration and exploitation of HF-RL-Course
llm_recipes
A set of scripts and notebooks on LLM finetunning and dataset creation
ayulockin.github.io
Personal website of Ayush Thakur 👉
BEAT
A Large-Scale Semantic and Emotional Multi-Modal Dataset for Conversational Gestures Synthesis [ECCV 2022]
cheese
Used for adaptive human in the loop evaluation of language and embedding models.
fashion-mnist-keras
Trying to push the accuracy on FashionMNIST dataset without using pre-trained weights and by training for 20 epochs only.
helm
Holistic Evaluation of Language Models (HELM), a framework to increase the transparency of language models (https://arxiv.org/abs/2211.09110). This framework is also used to evaluate text-to-image models in Holistic Evaluation of Text-to-Image Models (HEIM) (https://arxiv.org/abs/2311.04287).
llama_index
LlamaIndex (GPT Index) is a data framework for your LLM applications
lm-evaluation-harness
A framework for few-shot evaluation of autoregressive language models.
lmql
A query language for programming (large) language models.
nanoGPT-TF
Minimal implementation of GPT-2 in TensorFlow
odsc-yolo-e2e
Supplementary source code for our half day training session at ODSC 23 on end to end MLOps
pytorch-lightning
Pretrain, finetune and deploy AI models on multiple GPUs, TPUs with zero code changes.
SageMakerStudio
A repo showcasing SMSL and W&B
transformers
🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
trlx
A repo for distributed training of language models with Reinforcement Learning via Human Feedback (RLHF)
wandb-addons
Weights & Biases Addons is a repository consisting of additional unitilities and community contributions for supercharging your Weights & Biases workflows.
yolov5
YOLOv5 🚀 in PyTorch > ONNX > CoreML > TFLite