trannhiem Goto Github PK

aitemplate

AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (NVIDIA GPU) and MatrixCore (AMD GPU) inference.

autonomous_driving_visual_instruction_dataengine

Large Scale Quality -- Quantity -- Disversity Visual Instruction Dataset for Autonomous Driving

awesome-self-supervised-papers

Paper bank for Self-Supervised Learning

boostraping_multi_modalities_ssl

This repository collectes multiple state-of-the-art vision transformer model for accelerating vision extractor in multi-modalities model training.

building_torch_lightning_model

This repo will contain new adding feature and testing development project

codeformer

[NeurIPS 2022] Towards Robust Blind Face Restoration with Codebook Lookup Transformer

compression_machine_learning_tinyml

diffusers

🤗 Diffusers: State-of-the-art diffusion models for image and audio generation in PyTorch

distributed_training_single_and_multi_machines

foxbrain_llms

項目內容包括：1. 創建多樣化、高質量的中文教學數據集。2. 在開源語言模型（如bloomz、LLaMA2、Yi, Qwen等）上進行LLM訓練、微調、評估和測試。Building a diverse and high-quality Chinese instruction dataset. 2. LLM training, finetuning, evaluating, and testing on open-source language models

galatica

grit

GRiT: A Generative Region-to-text Transformer for Object Understanding (https://arxiv.org/abs/2212.00280)

hackathon_nlp_mvqg

This repository use for MVQA

hapiclr

Heuristic Attention Pixel-level for Self-supervised Contrastive Representation Learning

heuristic_attention_represenation_learning_ssl_pytorch

Heuristic Attention Representation Learning for Self-Supervised Pre-training

heuristic_attention_representation_learning_ssl_tensorflow

Heuristic Attention Representation Learning for Self-Supervised Pre-training

llm_scientific_paper

Model API for GALACTICA

ma_ssl

solo-learn: a library of self-supervised methods for visual representation learning powered by Pytorch Lightning

ma_ssrl_pytorch

MASSRL (Mulit-Augmentation Strategies on Searched and Random Augmentation Policies) Self-supervised visual representation learning.

ma_ssrl_tensorflow

min_dalle

min(DALL·E) is a fast, minimal port of DALL·E Mini to PyTorch

multimodal_integrated_app

This repository contains a web-based application that integrates speech, language, and visual understanding to provide a more intuitive and interactive user experience. The app uses state-of-the-art machine learning models and modern web technologies to analyze and understand speech, text, and images.

mup

maximal update parametrization (µP)

mv_ma_ssrl

Multi-View, Multi-Augmentation Self-Supervised Representation Learning

neural_net_architectures

Collection of Efficient Neural Net Architectures

neural_network_architecture_search

otter

🦦 Otter, a multi-modal model based on OpenFlamingo (open-sourced version of DeepMind's Flamingo), trained on MIMIC-IT and showcasing improved instruction-following and in-context learning ability.

self-supervised-learning-literature-survey

self_supervised_learning_demo

This repository evaluate the obtained representation learning from Self-Supervised Learning.

semi_self_supervised_project