Deep Learning in Image Processing Tutorial
1、Image classification
| Network Architecture | Code | Pre-training weight | PPT |
|---|---|---|---|
| ☑ AlexNet | Code | Google Drive | |
| ☑ VGG | Code | Google Drive | |
| ☑ ResNet | Code | Google Drive | |
| ☑ custom_dataset | Code | ||
| ☑ ConfusionMatrix | Code | ||
| ☑ RecptiveField | |||
| ☑ analyze_weights_featuremap | Code | ||
| ☑ batch normalization ☑ layer normalization ☑ group normalization | Code | ||
| ☑ fvcore | Code | ||
| ☑ tensorboard | |||
| ☑ cosine learning rate | |||
| ☑ Transformer |
2、Machine Learning
| machine learning | |
|---|---|
| ☑newton | |
| ☑SGD/AdaGrad/Adam | |
| ☑PCA | PCA |
| ☑Singular value decompisition | |
| ☑Linear Algebra | Notes |
| ☐Matrix derivative |
3、Numpy
| numpy | |
|---|---|
| ☑numpyBroadcast |
4、Image Segmentation
5、Image Fusion
⚡Dependence environment
matplotlib==3.3.4
pandas==1.1.5
Pillow==9.0.1
prettytable==2.1.0
torch==1.9.0
torchvision==0.10.0
tqdm==4.61.2
We recommend using pipreqs（pip list pipreqs）, a tool that has the advantage of automatically discovering which libraries are used by scanning the project directory and automatically generating a list of dependencies.
How to use ?
- Use pipreqs in the root directory of the project . / （pipreqs ./ ）
⚡Problems
- If the formulas in the MD file are not displayed correctly, please download the plugin MathJax 3 Plugin for Github.