a154609 Goto Github PK

pytorch-image-models

PyTorch image models, scripts, pretrained weights -- ResNet, ResNeXT, EfficientNet, EfficientNetV2, NFNet, Vision Transformer, MixNet, MobileNet-V3/V2, RegNet, DPN, CSPNet, and more

pytorch-lightning

The lightweight PyTorch wrapper for high-performance AI research. Scale your models, not the boilerplate.

pytorch-metrics

Accuracy, precision, recall, confusion matrix computation with batch updates

pytorch-opcounter

Count the MACs / FLOPs of your PyTorch model.

pytorch-yolov3

Minimal PyTorch implementation of YOLOv3

pytorch_classification

利用pytorch实现图像分类的一个完整的代码，训练，预测，TTA，模型融合，模型部署，cnn提取特征，svm或者随机森林等进行分类，模型蒸馏，一个完整的代码

pytorch_simple_centernet_45

A simple pytorch implementation of CenterNet (Objects as Points)

repvgg

RepVGG: Making VGG-style ConvNets Great Again

scaledyolov4

segdec-net-jim2019

Surface Defect Detection with Segmentation-Decision Network on KolektorSDD

senet

Squeeze-and-Excitation Networks

shufflenet

ShuffleNet in PyTorch. Based on https://arxiv.org/abs/1707.01083

simple-retinanet-pytorch

Rtinanet implement in pytorch.

smallobjectaugmentation

Simple implementation of augmentation for small object detection

ssd

High quality, fast, modular reference implementation of SSD in PyTorch

ssd-pytorch

基于pytorch版ssd进行改进注入CBAM空间通道注意力机制，加入FPN，类别损失函数改为focalloss

ssd_keras

A Keras port of Single Shot MultiBox Detector

surface-defect-detection

缺陷检测文献记录

tensorflow_exercises

The codes I made while I practiced various TensorFlow examples

text_gcn

Graph Convolutional Networks for Text Classification. AAAI 2019

tnt

Simple tools for logging and visualizing, loading and training

torchsummaryx

torchsummaryX: Improved visualization tool of torchsummary

transferlearning

Everything about Transfer Learning and Domain Adaptation--迁移学习

voc2007_for_yolo_torch

prepare dataset for voc https://github.com/ultralytics/yolov3

x-sdd-a-new-benchmark

yolo-fastest

:zap: Yolo universal target detection model combined with EfficientNet-lite, the calculation amount is only 230Mflops(0.23Bflops), and the model size is 1.3MB

yolo-reproduce-summary

YOLO reproduce summary (now based on YOLOv3)

yolov3

YOLOv3 in PyTorch > ONNX > CoreML > iOS

yolov4-tiny-keras

这是一个YoloV4-tiny-keras的源码，可以用于训练自己的模型。

yolov5

YOLOv5 in PyTorch > ONNX > CoreML > TFLite