One-click Deployment of Llama3 8B/70B! Only Occupies 1.07GB of Storage at Most; COCONut Is Launched, the First Large-scale Panoramic Image Segmentation Dataset Launched by ByteDance
Recently, Meta shocked the world by releasing the large model Llama3, which is known as the "open source GPT-4". This model has shocked the entire AI circle, and HyperAI can't wait to experience it!The hyper.ai official website has now launched the public tutorial "One-click deployment of Llama3"!Is anyone rubbing their hands as excitedly as I am? Come and run it~
From April 22nd to April 26th, hyper.ai official website updated quickly:
* High-quality public datasets: 10
* Selection of high-quality tutorials: 3
* Community article selection: 5 articles
* Popular encyclopedia entries: 5
Visit the official website:hyper.ai
Selected public datasets
1. COCONut Large-Scale Panoramic Image Segmentation Dataset
COCONut is the first large-scale manually annotated panoramic image segmentation dataset released by ByteDance, containing about 383K images and 5.18 million manually annotated panoramic segmentation masks. This achievement has been selected for CVPR2024.
Direct use:https://go.hyper.ai/43gIy
2. Solar System Planets Solar System Planets Dataset
The dataset contains .jpg images of all the planets in the solar system. The dataset is divided into training set, test set and validation set. Each set contains subsets of 9 different planets: Earth, Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus and Neptune.
Direct use:https://go.hyper.ai/awnIo
3. Chinese medical dialogue data
This dataset is a Chinese medical question-and-answer dataset, which contains question-and-answer data folders of 6 different medical departments, each of which contains a csv file.
Direct use: https://go.hyper.ai/scsmp
4. TAL-SCQ5K TAL Chinese Mathematics Competition Dataset
TAL-SCQ5K is a set of high-quality Chinese mathematics competition datasets, including 5K Chinese mathematics competition questions (3K for training and 2K for testing), available in Chinese and English.
Direct use: https://go.hyper.ai/RqwbX
5. Anhui Telecom Question and Answer Dataset
This dataset contains 156,000 Anhui Telecom question and answer data, including user questions, netizen answers, and best answers. The dataset comes from Baidu Knows and is suitable for FAQ question and answer system tasks.
Direct use: https://go.hyper.ai/CMkpM
6. WikiArt Wiki Art Image Dataset
The dataset contains 80,020 unique images from 1,119 different artists and 27 styles. The data comes from WikiArt.org. The WikiArt dataset can only be used for non-commercial research purposes.
Direct use: https://go.hyper.ai/nCKMw
This dataset contains 50 of the most amazing landmarks in the world, selected from the famous New Seven Wonders list. The dataset information includes the landmark's: name, type, location, Wikipedia link, image link, and year of construction.
Direct use: https://go.hyper.ai/qQLA4
8. Google Landmarks Photos Google Landmarks Image Dataset
The dataset contains URLs of about 120,000 landmark images, each of which has a unique ID, which helps with instance recognition and image retrieval tasks.
Direct use:https://go.hyper.ai/fJ5kO
This dataset is a shape dataset of world maps, including maps of various scales, such as national borders, first-level administrative management (provinces, states, etc.), disputed areas, coastlines, and densely populated areas.
Direct use:https://go.hyper.ai/hcwKe
10. SSDO Large-Scale Sea Surface Dissolved Oxygen Dataset
This dataset is the data used in the paper "Satellite Global Sea Surface Oxygen Mapping and Interpretation Based on Spatiotemporal Machine Learning". Based on the framework DOsurface-Pred Framework built in the paper, the research team generated a large-scale sea surface dissolved oxygen dataset spanning 2010-2018, including satellite data, fixed ships and conductivity meter measurement data.
Direct use:https://go.hyper.ai/d9w8B
For more public datasets, please visit:
Selected Public Tutorials
1. Deploy Llama3-8B-Instruct using Ollama and Open WebUI
This tutorial is a one-click run package of Ollama + Open WebUI. You only need to enter commands step by step to run the Llama3-8B-Instruct large model.
Run online:https://go.hyper.ai/7lZyb
2. Deploy Llama3-70B using Ollama and Open WebUI
This tutorial uses Ollama and Open WebUI tools to run the model with one click. You only need to follow the steps and enter commands to run the Llama3-70B large model. Although the model parameters are large, it can be used with a single A6000 card after platform deployment, occupying only 1.07 GB of storage.
Run online:https://go.hyper.ai/u4XHj
As a universal and stable real-time audio-driven 3D talking face generation technology, GeneFace++ can achieve real-time talking human generation by improving lip synchronization, video quality and system efficiency. The "GeneFace++ Digital Human Demo" tutorial is now available on the hyper.ai official website, and the environment has been set up for everyone, simplifying the production process of digital humans. There is no need to worry about environment configuration, hardware requirements and version compatibility. Click on the online run to start it with one click.
Run online:https://go.hyper.ai/4BwgN
Community Articles
Here comes the useful information! hyper.ai has compiled 10 mathematical classification datasets, including dataset resources highly recommended by genius mathematician Terence Tao, to help the development and research of AI for Math.
View detailed information:https://go.hyper.ai/6Vhrb
Based on the rapid development of AI, the Google team studied how machine learning models can effectively convey research results to radiologists, evaluated the impact of generalized AI-assisted systems on lung cancer screening workflows under specific workflow environments, equipment, and country-specific guidelines and scoring/management protocols, and conducted tests in the United States and Japan. The results showed that with the assistance of AI, the average screening time per case can be reduced by 14 seconds. The research paper has been published in the journal "Radiology AI".
View the full report:https://go.hyper.ai/wiCl0
HyperAI has collected AI for Science conferences at home and abroad, and organized conference information and links. You can go directly to the official homepage with one click, so you don’t miss any important events! This article is a summary of recent events.
View the full article:https://go.hyper.ai/HTGNG
Professor Fang Ning's team at Xiamen University has developed an automated, high-speed, multi-dimensional single-particle tracking system based on deep learning, which breaks the limitation of nanoparticle rotation tracking in cell microenvironments and achieves all-round and accurate tracking of single molecules/single nanoparticles in living cells at the nanoscale. This article is a detailed interpretation and sharing of the research. The paper has been published in the authoritative journal "Nano Letters".
View the full report:https://go.hyper.ai/G4k2e
Researchers from the Zhejiang University GIS Laboratory combined machine learning technology with satellite products to develop a global sea surface dissolved oxygen comprehensive modeling framework DOsurface-Pred Framework, which confirmed that even in oxygen-rich sea surface areas, the dissolved oxygen content has shown a downward trend, and this trend is mainly attributed to changes in sea surface temperature. The research paper has been published in "ACS Publications".
View the full report:https://go.hyper.ai/NlRmn
Popular Encyclopedia Articles
1. Glitch Token
2. Multimodal Large Language Model MLLM
3. Masked Language Modeling (MLM)
4. Group Query Attention GQA
5. Variational Autoencoder VAE
Here are hundreds of AI-related terms compiled to help you understand "artificial intelligence" here:
Station B live broadcast preview
Google recently announced that it will hold the 2024 I/O Developer Conference on May 14th local time. In order to help everyone gain a deeper understanding,The Super Neuron Live Room will continue to broadcast the "Google Special" video.Involves: Google I/O press conferences over the years, interviews with executives, related documentaries and other rich content.
The following table is a preview of the content selected by the editor↓↓↓
| date | time | content |
| April 29 Monday | 18:00 | Google I/O Conferences over the Years |
| Tuesday, April 30 | 18:00 | Google Cloud NEXT Conferences |
| Wednesday, May 1 | 18:00 | TIME100 Interview with Sundar Pichai |
| Thursday, May 2 | 18:00 | Google CEO on the US-China AI race |
| Friday, May 3 | 18:00 | AlphaGo Documentary |
| Saturday, May 4 | 18:00 | The story behind the founder of Google |
| Sunday, May 5 | 18:00 | BBC documentary: A World Without Google |
Super Neuro TV broadcasts live 24/7. Click to get the "electronic pickles" in the AI field:
http://live.bilibili.com/26483094
The above is all the content of this week’s editor’s selection. If you have resources that you want to include on the hyper.ai official website, you are also welcome to leave a message or submit an article to tell us!
See you next week!
About HyperAI
HyperAI (hyper.ai) is the leading artificial intelligence and high-performance computing community in China.We are committed to becoming the infrastructure in the field of data science in China and providing rich and high-quality public resources for domestic developers. So far, we have:
* Provide domestic accelerated download nodes for 1400+ public data sets
* Includes 300+ classic and popular online tutorials
* Interpretation of 100+ AI4Science paper cases
* Support 500+ related terms search
* Hosting the first complete Apache TVM Chinese documentation in China
Visit the official website to start your learning journey: