Welcome!

I am a computer vision deep learning developer working in Korea.
This is my blog, and you can see everything I've studied here. https://www.notion.so/pervin0527

Repository configuration

Source

The codes here are generally required to use Image Classification or Object Detection API.
If there are any improvements to the code, please feel free to write them down in the Issues.

1. Image Data Augmentations

• Albumentations, pandas, matplotlib, opencv-python, etc.
• Dataset augmentation for Image Classification
• Dataset augmentation for Object Detection
• Cutmix & Mixup augmentation

2. Image Classification

• Tensorflow, Keras, scikit-learn, pandas
• Multi Class Image Classification
• Train with Multi GPUs
• tf.data API + augmentations
• Multi Label Image Classification
• K-Fold Cross Validation & Ensemble

3. Object Detection

• Tensorflow 2 Object Detection API
• Yolo v4
• Custom SSD_ResNetV2
• Quantization Int8
• EfficientDet lite

4. Competitions

• [Dacon] landmark image classification (26th / 387teams)
• [Dacon] Dirty Mnist multi label classification (10th / 780teams)
• [ai connect] 3D pose estimation (7th / 35teams)

5. 3D

• PointNet
• PointCutMix

DL_Note

It summarizes the details of machine learning and deep learning that are easier to understand.

1. Simple Linear Regression
2. How to minimize cost?
3. Multi variable linear regression
4. Logistic Regression
5. Softmax
6. learning rate decay, overfitting
7. Dataset, Learning method
8. Neural Network, XOR gate
9. ReLU activation func
10. weight initialization, dropout, batch normalization
11. Convolutional Neural Network
12. Recurrent Neural Network