# 0501 Machine Learning Tutorial

### 01 環境設定

1. 安裝 Platform
* (Windows)[Anaconda](https://www.anaconda.com/)
* (Linux) [Docker](https://www.docker.com/)
2. 安裝 Nvidia SDK
[NVidia Cuda Toolkit 11.1/11.2](https://developer.nvidia.com/cuda-11.1.1-download-archive?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal)
3. 安裝Editor
* [Visual Studio Code](https://code.visualstudio.com/)
* Jupyter

pip install juypter

* Spyder
* Pycharm
4. 建立虛擬環境(Virutal Envirnment)

conda create -n $envname python=3.8

$envname 為虛擬環境名稱 : test

conda activate test

### 02 安裝Framework

1. [tensorflow](https://www.tensorflow.org/install)
2. [Pytorch](https://pytorch.org/get-started/locally/)

版本建議安裝舊版 1.9.0

pip install torch==1.9.0+cu111 torchvision==0.10.0+cu111 torchaudio==0.9.0 -f https://download.pytorch.org/whl/torch_stable.html

### 03 測試環境

[TIMM Models Resnet](https://rwightman.github.io/pytorch-image-models/models/resnet/)

### 04 Training

[Transfer Learning](https://www.pluralsight.com/guides/introduction-to-resnet)

### Bookmarks

1. PaperwithCodes(https://paperswithcode.com/)
2. ModelZoo
* [pytorch-image-models](https://github.com/rwightman/pytorch-image-models)
* [torchvision](https://pytorch.org/vision/stable/index.html)
* [onnx models](https://github.com/onnx/models)
* [openvino model zoo](https://github.com/openvinotoolkit/open_model_zoo)
3. Youtuber
* [李弘毅](https://www.youtube.com/c/HungyiLeeNTU)
* 林軒田
4. Tutorials
* [Pytorch handbook](https://handbook.pytorch.wiki/chapter1/1.2-pytorch-installation.html)

5. [Kaggle](https://www.kaggle.com/)
6. [netron](https://netron.app/)