생성형 AI

deeplearning4j

• https://github.com/deeplearning4j/

  • https://github.com/clojure-interop/deeplearning4j

• Attention Is All You Need

  • https://blog.research.google/2017/08/transformer-novel-neural-network.html

• Javaでディープラーニングをするためのファーストステップ

• 머신러닝 - 데이터주도

  • 딥러닝 - 뉴럴 네트워크
    • 생성형ai

<혼자 공부하는 머신러닝+딥러닝>의 코드 저장소입니다. https://github.com/rickiepark/hg-mldl

https://namu.wiki/w/%EB%94%A5%20%EB%9F%AC%EB%8B%9D 밑바닥부터 시작하는 딥러닝 1,2,3,4 https://www.coursera.org/learn/neural-networks-deep-learning 모두를 위한 머신러닝/딥러닝 강의

• 모두를 위한 딥러닝 강좌 시즌 1

Neural Networks

• Neural networks are function approximators that stack affine transformations followed by nonlinear transformations.

Linear Neural Networks

NPL(Natural Language Processing)

• 자연어 처리

etc

github cloud Llama 2 code llama-7a/34b mistral-7b

ollama

• https://github.com/jmorganca/ollama

• https://ollama.ai/

• Running Ollama on Windows ! Ubuntu Version | (Much Awaited Video)

  • Step 1: Turn Windows Features on or off (Virtual Machine Platform and Windows System for Linux)
  • Step 2: Go to Command Prompt and say “wsl --install”
  • Step 3: In the Command Prompt type “wsl --user root -d ubuntu”
  • Step 4: Type curl https://ollama.ai/install.sh | sh
  • ollama serve
  • ollama run llama2

• ui

  • https://github.com/ollama-webui/ollama-webui
    • git clone https://github.com/ollama-webui/ollama-webui.git && cd ollama-webui/
    • cp -RPp example.env .env
    • https://github.com/nvm-sh/nvm
      • nvm install 16
      • nvm use 16
    • npm i && npm run build
    • cd ./backend
      • sudo apt install python3-pip
    • pip install -r requirements.txt
      • pip install uvicorn
    • sh start.sh
      • python3 -m uvicorn main:app
    • http://localhost:8000
      • Model not selected??
    • cd ~/ollama-webui/backend && python3 -m uvicorn main:app
  • https://github.com/oobabooga/text-generation-webui.git

• taskweaver

  • https://microsoft.github.io/TaskWeaver/docs/llms/ollama

• Google Colab

  • https://colab.research.google.com/
  • https://research.google.com/colaboratory/faq.html?hl=ko

• mojo

  • https://www.modular.com/max/mojo
  • https://github.com/modularml/mojo