Skip to content

Notes, Codes, and Tutorials for the Deep Learning Course at ChinaHadoop

License

Notifications You must be signed in to change notification settings

liutaobit/DeepLearningCourseCodes

 
 

Repository files navigation

Deep Learning Course Codes

Notes, Codes, and Tutorials for the Deep Learning Course at ChinaHadoop

注意每一份代码分别有Jupyter Notebook, Python, 以及HTML三种形式,大家可以按照自己的需求阅读,学习或运行。 运行时需要注意anaconda的版本问题,anaconda2-5.0.0与anaconda3-5.0.0分别对应python2.7与python3.6环境。

重要参考资料:

  1. Stanford CS229 Machine Learning, Fall 2017
  2. Deep Learning Book读书笔记

学习资料:

  1. Effective TensorFlow - TensorFlow tutorials and best practices.
  2. Finch - Many Machine Intelligence models implemented (mainly tensorflow, sometimes pytorch / mxnet)
  3. Pytorch Tutorials - PyTorch Tutorial for Deep Learning Researchers.
  4. MXNet the straight dope - An interactive book on deep learning. Much easy, so MXNet. Wow.

第一讲:深度学习课程总览与神经网络入门

代码示例:TensorFlow基础与线性回归模型(TensorFlow, PyTorch)

第二讲:传统神经网络

代码示例:K近邻算法,线性分类,以及多层神经网络(TensorFlow, PyTorch)

第三讲:卷积神经网络基础

代码示例:卷积神经网络的基础实现(TensorFlow)

第四讲:卷积神经网络进阶

代码示例:卷积神经网络的进阶实现(TensorFlow)

第五讲:深度神经网络:目标分类与识别

代码示例:深度神经网络-图像识别与分类(TensorFlow, PyTorch)

pip install git+https://github.com/zsdonghao/tensorlayer.git
conda install -c menpo opencv3 
或
pip install opencv-python
  • 所需数据集下载:data.zip: [微云][百度云] (覆盖./05_Image_recognition_and_classification/data文件夹)  
  • 所需模型下载: vgg19.npz  [微云][百度云] (放置于./05_Image_recognition_and_classification文件夹下)  
  • 所需模型下载:inception_v3.ckpt [微云][百度云] (放置于./05_Image_recognition_and_classification文件夹下)

第六讲:深度神经网络:目标检测与定位

代码示例:目标检测模型示例 (TensorFlow, PyTorch)

  1. TensorFlow Object Detection API使用示例
  • 所需模型下载:ssd_mobilenet_v1_coco_11_06_2017: [微云] (解压并置于06_Object_detection/Object_Detection_Tensorflow_API_demo/object_detection/文件夹下)
  1. [SSD: Single Shot Multibox Detector] (TensorFlow实现, PyTorch实现)

  2. [YOLO, YOLOv2] (TensorFlow实现, PyTorch实现)

第七讲:深度神经网络:目标追踪与目标分割

代码示例:目标追踪与目标分割

  1. 目标追踪 - [GOTURN](TensorFlow实现, Plain Python实现, Original C++实现)

  2. 目标分割 - [FCN](TensorFlow实现, PyTorch实现)

  3. 目标分割 - [Mask-RCNN](TensorFlow实现, PyTorch实现)

第八讲:循环神经网络与序列模型

第九讲:无监督式学习

第十讲:增强学习

About

Notes, Codes, and Tutorials for the Deep Learning Course at ChinaHadoop

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 60.4%
  • HTML 25.4%
  • Python 14.2%