We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
删除线表示任务完成或基本完成。
The text was updated successfully, but these errors were encountered:
感谢作者的分享~不知作者有没有增加HRNet相关model的计划https://github.com/HRNet/HRNet-Semantic-Segmentation?
Sorry, something went wrong.
No branches or pull requests
删除线表示任务完成或基本完成。
参考schedulers和optimizers
参考StepLR
增加模型DeConvolution,参考 Learning Deconvolution Network for Semantic Segmentation #48增加模型FRRN A和FRRN B,参考 Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes #50增加GPU资源不足时用梯度累加近似多batch训练,参考博客PyTorch 大批量数据在单个或多个 GPU 训练指南,仍在调试测试性能,测试没有问题,但是训练需要注意,这种操作相当于增加了batch,需要设置相当的学习率,原先batch size为1时学习率设置较小1e-4,如果用了grad_acc_steps=5那么可以设置稍大的学习率1e-5阅读Wider or Deeper: Revisiting the ResNet Model for Visual Recognition论文并实现相关结构,参考 Wider or deeper: Revisiting the resnet model for visual recognition #15 ,下一步实现Wider or Deeper分类网络和对应的分割网络阅读Large Kernel Matters Improve Semantic Segmentation by Global Convolutional Network并实现相关网络结构,参考 Large Kernel Matters -- Improve Semantic Segmentation by Global Convolutional Network #51阅读Context Encoding for Semantic Segmentation论文并实现相关结构,参考 Context Encoding for Semantic Segmentation #52The text was updated successfully, but these errors were encountered: