PaddleSeg
PaddleSeg copied to clipboard
Easy-to-use image segmentation library with awesome pre-trained model zoo, supporting wide-range of practical tasks in Semantic Segmentation, Interactive Segmentation, Panoptic Segmentation, Image Mat...
Hi, while I train , and do validation I get the validation accuracy and the training loss, those two are not enough to determine if my model overfits or not...
模型训练
请问edgeflow的模型训练、网络搭建部分的算法在哪里呢? --- 欢迎您反馈PaddleSeg使用问题,辛苦您提供以下信息,方便我们快速定位和解决问题: 1. PaddleSeg版本:(请提供版本号和分支信息,如PaddleSeg release/2.3) 2. PaddlePaddle版本:(如PaddlePaddle 2.1.0) 3. 操作系统信息:(如Linux/Windows/MacOS) 4. Python版本号:(如Python3.6/7/8) 5. CUDA/cuDNN版本:( 如CUDA10.2/cuDNN 7.6.5等) 6. 其他内容: (增加其他与问题相关的内容)
求教,我想像pytorch一样加载Paddle2.0训练好的模型参数model.pdparams,运用Summary函数可以输出参与训练的参数量和总的模型参数量,请问该怎么写呢? The number of parameters involved in the training and the total number of parameters in the model can be displayed by using the Summary function. import paddle save_path =...
我在使用PSPNet运行自己的数据集,实在是没有找到为什么optimizer.step()这里会出现TypeError: 'str' object is not callable错误,我尝试着将Image和label的值都打印出来查看了,里面是有值的,计算的loss也是有值的,这可能是哪里出现了问题呢?希望可以得到帮助 注意:我没有改动任何和模型有关的文件,仅仅修改了配置相应的.yml文件和加载数据集的.py文件,下面是报错的信息:   这是我加载的数据集文件(已做修改)   配置文件pspnet.yml信息   期待您的回复,万分感谢!!!
paddleseg2.5 paddlePaddle 2.1.0 Linux Python3.7 cuda11.2 用VisionTransformer作为backbone,输入的图像宽高必须相等吗? 
Thanks for your issue. To help us solve the issue better, please provide following information: 1. PaddleSeg version: (please specify the branch as well,e.g. PaddleSeg release/2.3) 2. PaddlePaddle version: (e.g....
# CutMix图像增强 - [CutMix](https://arxiv.org/abs/1905.04899v2): Regularization Strategy to Train Strong Classifiers with Localizable Features - 在图像分割中使用CutMix图像增强,效果确实还挺好的,希望能在图像增强中增加这个功能,谢谢。
this bug happens due to the fact that there exists different types of data parsing in prepare.py in MedicalSeg.Here the 442th line is original code and the 441th line is...
Thanks for your bug report. To help us solve the issue better, please provide following information: 1. PaddleSeg version: (please specify the branch as well,e.g. PaddleSeg release/2.3) 2. PaddlePaddle version:...
https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.5/contrib/CityscapesSOTA Cityscapes的这个打榜模型有HRNet_W18_NV,HRNet_W32_NV和HRNet_W64_NV这几种不同大小backbone的预训练模型么,我看到conf里面用的是HRNet_W48_NV,然后readme里面也只给了HRNet_W48_NV的pretrain weights 这个是HRNet_W48_NV 的pretrain weights链接https://bj.bcebos.com/paddleseg/dygraph/cityscapes/ocrnet_hrnetw48_mapillary/pretrained.pdparams