Deep Learning (DL)
Yes you should understand backprop
if you’re using sigmoids or tanh non-linearities in your network and you understand backpropagation you should always be nervous about making sure that the initialization doesn’t cause them to be fully saturated.
z = 1/(1 + np.exp(-np.dot(W, x))) # forward pass
dx = np.dot(W.T, z*(1-z)) # backward pass: local gradient for x
dW = np.outer(z*(1-z), x) # backward pass: local gradient for W
z = np.maximum(0, np.dot(W, x)) # forward pass
dW = np.outer(z > 0, x) # backward pass: local gradient for W
- FLOPS与GOPS：各平台及神经网络算力算量调研
- 玩转 FLOPS – ARM 如何测量 GPU 计算性能
- how to calculate a net's FLOPs in CNN
- https://github.com/Lyken17/pytorch-OpCounter
- https://github.com/1adrianb/pytorch-estimate-flops
- How to understand / calculate FLOPs of the neural network model?
- 【YOLO学习】召回率（Recall），精确率（Precision），平均正确率（Average_precision(AP) ），交除并（Intersection-over-Union（IoU））
- Evaluation Metrics for Machine Learning - Accuracy, Precision, Recall, and F1 Defined
- Intersection over Union (IoU) for object detection
- https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance
- Calculating Variance: Welford's Algorithm vs NumPy
-
https://stanford.edu/~shervine/teaching/cs-230/cheatsheet-recurrent-neural-networks
-
Illustrated Guide to LSTM’s and GRU’s: A step by step explanation
-
onnx实现
-
tensorflow实现
- https://github.com/mjq11302010044/RRPN/blob/master/caffe-fast-rcnn/src/caffe/layers/rotate_roi_pooling_layer.cpp
- https://github.com/mjq11302010044/RRPN/blob/master/caffe-fast-rcnn/src/caffe/layers/rotate_roi_pooling_layer.cu
- https://github.com/onnx/onnx/blob/master/docs/Operators.md#Resize
- 是什么引起了各个框架 Resize 操作的结果不同？——来自 ONNX 的标准化尝试
- https://github.com/onnx/onnx/pull/3026
- https://github.com/onnx/onnx/blob/master/onnx/backend/test/case/node/resize.py
- http://www.xavierdupre.fr/app/mlprodict/helpsphinx/onnxops/onnx__ScatterND.html
- [TOPI][CUDA] “scatter_nd”’ has a very poor performance on CUDA backend (>1000x slower than hand-written cuda code)
- [技术干货] 【昇腾众智】RetinaFace_ResNet50模型的910训练+310推理
-
https://github.com/onnx/onnx/blob/main/docs/Operators.md#Einsum
-
Einsum is All you Need - Einstein Summation in Deep Learning
- Feedforward neural network
- What is a Feed Forward Neural Network?
- Deep Learning (Part 1) - Feedforward neural networks (FNN)
- 人脸检测——MTCNN
- https://github.com/Seanlinx/mtcnn
- https://github.com/kpzhang93/MTCNN_face_detection_alignment
- MTCNN人脸及特征点检测---代码应用详解（基于ncnn架构）
- https://github.com/hualitlc/MTCNN-on-FDDB-Dataset
- https://github.com/itlab-vision/FDDB/tree/master/fddb_evaluate
-
https://github.com/tanluren/yolov3-channel-and-layer-pruning
-
How to Implement a YOLO (v3) Object Detector from Scratch in PyTorch: Part 1
-
How to implement a YOLO (v3) object detector from scratch in PyTorch: Part 2
-
How to implement a YOLO (v3) object detector from scratch in PyTorch: Part 3
-
万字长文，用代码的思想讲解Yolo3算法实现原理，Visdrone数据集和自己制作数据集两种方式在Pytorch训练Yolo模型
- https://docs.ultralytics.com/models/yolo12/#key-features
- https://github.com/sunsmarterjie/yolov12
-
A Brief History of CNNs in Image Segmentation: From R-CNN to Mask R-CNN
-
How to calculate feature map after applying ROI Align, as explained in Mask RCNN Paper?