Машинное обучение и Большие данные: различия между версиями

Материал из Artem Aleksashkin's Wiki
Перейти к навигации Перейти к поиску
 
(не показано 78 промежуточных версий этого же участника)
Строка 1: Строка 1:
[[Файл:Ai-brain.jpg|400px]]
[[Файл:Ai-brain.jpg|400px]]


= Hardware =
= Software installation =


* Lenovo x230 + eGPU
* Anaconda - https://www.anaconda.com/products/individual
** [https://aliexpress.ru/item/32983647923.html Expresscard V8.0 EXP GDC Beast PCIe PCI-E]
** To remove (base) from PS1 - conda config --set changeps1 false
** Блок питания на 350-600 ватт
* sklearn - https://scikit-learn.org/stable/
** Nvidia GeForce 760 4gb
* CatBoost - https://catboost.ai/
** [https://egpu.io/forums/builds/thinkpad-x230-express-card-2-0-5-gt-s-windows-10-by-boelly/ Similar setup]
* LightGBM - https://lightgbm.readthedocs.io/en/latest/
** [https://egpu.io/forums/expresscard-mpcie-m-2-adapters/mpcieecngff-m2-resolving-detection-bootup-and-stability-problems/ Troubleshooting]
* XGBoost - https://xgboost.readthedocs.io/en/stable/
** 16 GB of mem will produce lags. Remove 1 stick of mem to 8 GB
* Tensorflow - https://www.tensorflow.org/install - [[Tensorflow for old GPUs]]
** Be sure that you GPU conneted to power fully(8+6 or 8+8) - it can produce 43 error
** [https://www.youtube.com/watch?v=p59MNoqWY9c eGPU setup Lenovo Thinkpad x230 with GTX 760 Part 1 ( setup )]
** [https://www.youtube.com/watch?v=xJsHLTCo9Ho eGPU setup Lenovo Thinkpad x230 with GTX 760 Part 2 ( Fixing Error 12)]
** [https://www.youtube.com/watch?v=qOoY30pubBg eGPU setup Lenovo Thinkpad x230 with GTX 760 Part 3 ( Gameplay )]
** In Windows go to Control panel, Hardware setup, Nvidia settings, 3d graphics, There you can select default video adapter
*** But it won't help to use GPU in games - you need to connect external screen to GPU and disable laptop screen. Then games will run on eGPU.


= Software =
<math>x^2+y^2=z^2</math>
* [https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html NVIDIA CUDA Installation Guide for Linux]
<pre>
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/cuda-ubuntu2004.pin
sudo mv cuda-ubuntu2004.pin /etc/apt/preferences.d/cuda-repository-pin-600
wget https://developer.download.nvidia.com/compute/cuda/11.2.2/local_installers/cuda-repo-ubuntu2004-11-2-local_11.2.2-460.32.03-1_amd64.deb
sudo dpkg -i cuda-repo-ubuntu2004-11-2-local_11.2.2-460.32.03-1_amd64.deb
sudo apt-key add /var/cuda-repo-ubuntu2004-11-2-local/7fa2af80.pub
sudo apt-get update
sudo apt-get -y install cuda
</pre>
* [https://www.tensorflow.org/install/gpu TensorFlow for GPU]
* [https://developer.nvidia.com/rdp/cudnn-download cuDNN SDK]
* [https://developer.nvidia.com/nvidia-tensorrt-7x-download TensorRT]
* [https://towardsdatascience.com/installing-tensorflow-gpu-in-ubuntu-20-04-4ee3ca4cb75d]


= Курсы =
= Курсы =
* [https://www.youtube.com/watch?v=tPYj3fFJGjk TensorFlow 2.0 Complete Course - Python Neural Networks for Beginners Tutorial]
* https://medium.com/nuances-of-programming/%D1%82%D0%BE%D0%BF-10-%D0%BA%D1%83%D1%80%D1%81%D0%BE%D0%B2-%D0%BF%D0%BE-%D0%BC%D0%B0%D1%88%D0%B8%D0%BD%D0%BD%D0%BE%D0%BC%D1%83-%D0%B8-%D0%B3%D0%BB%D1%83%D0%B1%D0%BE%D0%BA%D0%BE%D0%BC%D1%83-%D0%BE%D0%B1%D1%83%D1%87%D0%B5%D0%BD%D0%B8%D1%8E-%D0%B2-2020-1e1d870a24b7
* https://medium.com/nuances-of-programming/%D1%82%D0%BE%D0%BF-10-%D0%BA%D1%83%D1%80%D1%81%D0%BE%D0%B2-%D0%BF%D0%BE-%D0%BC%D0%B0%D1%88%D0%B8%D0%BD%D0%BD%D0%BE%D0%BC%D1%83-%D0%B8-%D0%B3%D0%BB%D1%83%D0%B1%D0%BE%D0%BA%D0%BE%D0%BC%D1%83-%D0%BE%D0%B1%D1%83%D1%87%D0%B5%D0%BD%D0%B8%D1%8E-%D0%B2-2020-1e1d870a24b7
* https://skillbox.ru/course/profession-data-scientist/
* https://skillbox.ru/course/profession-data-scientist/
Строка 69: Строка 51:
* SciPy - https://www.scipy.org/
* SciPy - https://www.scipy.org/
* Pandas - https://pandas.pydata.org/
* Pandas - https://pandas.pydata.org/
** [https://pandas.pydata.org/docs/getting_started/comparison/comparison_with_sql.html Pandas like SQL]
* Scikit-learn - https://scikit-learn.org/stable/
* Scikit-learn - https://scikit-learn.org/stable/
* Matplotlib - https://matplotlib.org/
* Matplotlib - https://matplotlib.org/
Строка 90: Строка 73:
* https://github.com/philipperemy/timit
* https://github.com/philipperemy/timit
* https://www.nist.gov/programs-projects/face-recognition-grand-challenge-frgc
* https://www.nist.gov/programs-projects/face-recognition-grand-challenge-frgc
= NLP =
* https://ru.wikipedia.org/wiki/GPT-3
* https://russiannlp.github.io/rugpt-demo/
* https://copy.ai
* https://github.com/sberbank-ai/ru-gpts


= Железо и драйверы =
= Железо и драйверы =
Строка 120: Строка 110:


* https://medium.com/stereopi/opencv-and-depth-map-on-stereopi-tutorial-62cb6792bbed
* https://medium.com/stereopi/opencv-and-depth-map-on-stereopi-tutorial-62cb6792bbed
= Anaconda =
* https://www.anaconda.com/products/individual
* https://docs.anaconda.com/anaconda/install/linux/
* https://mas-dse.github.io/startup/anaconda-ubuntu-install/
= TensorFlow =
<embedvideo service="youtube" dimensions="800x450">https://www.youtube.com/watch?v=vfyZf2Wj3pU&list=PLA0M1Bcd0w8ynD1umfubKq1OBYRXhXkmH</embedvideo>
<embedvideo service="youtube" dimensions="800x450">https://www.youtube.com/watch?v=tPYj3fFJGjk</embedvideo>


= Некоторые полезные ресурсы =
= Некоторые полезные ресурсы =
Строка 145: Строка 147:
* https://arxiv.org/list/cs.CV/recent
* https://arxiv.org/list/cs.CV/recent
* https://yandex.ru/dev/catboost/
* https://yandex.ru/dev/catboost/
* [[Теория вероятностей]]

Текущая версия от 23:36, 19 октября 2023

Ai-brain.jpg

Software installation

Курсы

Большие данные

Методы

  • Теорема Байеса
  • Функции ошибки и регуляризация
  • Расстояние Кульбака-Лейблера и перекрестная энтропия
  • Градиентный спуск: основы
  • Граф вычислений и дифференцирование на нем
  • Перцептрон
  • Глубокие нейронные сети
  • Классификация
  • Кластеризация
  • Регрессия
  • Машинное зрение
  • Метод к-средних
  • word2vec

Библиотеки

Датасеты

NLP

Железо и драйверы

Темы

Face Recognition

Speech Recognition

Image Object Recognition

Anomaly Detection

Prediction

StereoVision

Anaconda

TensorFlow

Некоторые полезные ресурсы