Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / lxtgh/sfsegnets issues and pull requests
#81 - SFNet_Lite模型并没有找到
Issue -
State: open - Opened by CharlieFengCN 2 months ago
- 1 comment
#80 - 使用训练好的模型也需要mapillary数据吗
Issue -
State: open - Opened by kll-df 7 months ago
#79 - 请问怎样测试速度
Issue -
State: open - Opened by takaniwa 10 months ago
- 1 comment
#78 - SFNet-Lite DSN issue [Not using dsn in forward pass]
Issue -
State: closed - Opened by jnisarg 10 months ago
- 1 comment
#77 - 您好,请问为什么SRNet的模块ModuleHead, ChannelReasonModule的代码不在呢?我很想应用一下,请问您可以释放一下这部分的代码吗?
Issue -
State: closed - Opened by takaniwa about 1 year ago
- 1 comment
#76 - Trying to resize storage that is not resizable
Issue -
State: closed - Opened by ygyleo about 1 year ago
#75 - SenseTime's reproduction
Issue -
State: closed - Opened by sudeqian over 1 year ago
- 1 comment
#74 - FileNotFoundError: [Errno 2] No such file or directory: './data/mapillary/config.json'
Issue -
State: closed - Opened by jingyanliao over 1 year ago
- 1 comment
#73 - 请问训练的时候是加载单张图片进入模型吗?
Issue -
State: closed - Opened by qubewx almost 2 years ago
- 2 comments
#72 - Does anybody successfully convert this model to TensorRT?
Issue -
State: closed - Opened by dlz620301 almost 2 years ago
- 5 comments
#71 - “对齐”问题求教
Issue -
State: closed - Opened by ChengSongyang almost 2 years ago
- 1 comment
#70 - 请问怎么理解“对齐”
Issue -
State: closed - Opened by ChengSongyang almost 2 years ago
- 2 comments
#69 - SFNet-Lite: Gated Dual Flow Alignment Module
Issue -
State: closed - Opened by yongjin120 about 2 years ago
- 3 comments
#68 - How can I find the code about"Towards Efficient Scene Understanding via Squeeze Reasoning"
Issue -
State: closed - Opened by Yiir-1 about 2 years ago
- 2 comments
#67 - 请问为啥sfnet-lite代码和论文结构图不一致?
Issue -
State: closed - Opened by swjtulinxi about 2 years ago
- 4 comments
#66 - No module named "UperNetAlignHeadAddition" in file "network/sfnet_stdcnet.py"
Issue -
State: closed - Opened by JianmingLi-888 over 2 years ago
- 1 comment
#65 - GPU not being used in eval Cityscapes?
Issue -
State: closed - Opened by cuizy3 over 2 years ago
- 1 comment
#64 - Gridsampler to TensorRT
Issue -
State: closed - Opened by jyang68sh over 2 years ago
- 3 comments
#63 - Why did you remove the Resnet-101 Cityscape weights ?
Issue -
State: open - Opened by GaetanBaert over 2 years ago
- 2 comments
Labels: enhancement
#62 - Thanks for your great work and I need some help about how to resume training from the latest checkpoint?
Issue -
State: closed - Opened by alanlyawen over 2 years ago
#61 - --arch selected_architecture
Issue -
State: closed - Opened by m828 over 2 years ago
- 1 comment
#60 - Train on a single GPU
Issue -
State: closed - Opened by nur-917Hub over 2 years ago
- 1 comment
#59 - Train on less than 8 GPUs
Issue -
State: closed - Opened by BabaVegato over 2 years ago
- 1 comment
#58 - mapillary-vistas-dataset_public_v1.1.zip
Issue -
State: closed - Opened by wmnj almost 3 years ago
- 1 comment
#57 - pretrain for resnet18
Issue -
State: closed - Opened by jyang68sh almost 3 years ago
- 1 comment
#56 - nan for mIOU
Issue -
State: closed - Opened by niloofarAzari almost 3 years ago
- 3 comments
Labels: bug
#55 - about config.json
Issue -
State: closed - Opened by zhouzhennan about 3 years ago
- 1 comment
#54 - different classes lie in different groups along channels sparsely
Issue -
State: closed - Opened by whuhit about 3 years ago
- 3 comments
#53 - The "FAM" is indeed a stripped-down version of "DCN"
Issue -
State: closed - Opened by heziliustc about 3 years ago
- 2 comments
#52 - C盘的占用率是100%
Issue -
State: closed - Opened by gaowq2017 about 3 years ago
- 1 comment
#51 - Test mIoUs are better than val mIoUs
Issue -
State: closed - Opened by ShihuaHuang95 over 3 years ago
- 1 comment
#50 - Papers with Code Leaderboard
Issue -
State: closed - Opened by RolandGao over 3 years ago
- 1 comment
#49 - should i use multigrid[4, 8 16] in SRNet ? the paper use it ,but the code don't use?
Issue -
State: closed - Opened by swjtulinxi over 3 years ago
- 1 comment
#48 - 请问设个参数class_uniform_pct是什么意思?
Issue -
State: closed - Opened by swjtulinxi over 3 years ago
- 1 comment
#47 - Question about the flow_make and flow_warp functions.
Issue -
State: closed - Opened by wuwenbin970731 over 3 years ago
- 1 comment
#46 - Dilated ResNet in SFNet?
Issue -
State: closed - Opened by wondervictor over 3 years ago
- 4 comments
#45 - export sfnet_r18_78.pth to onnx
Issue -
State: closed - Opened by fjremnav over 3 years ago
- 1 comment
#44 - License?
Issue -
State: closed - Opened by Jason93K over 3 years ago
- 2 comments
#43 - there is not DeepR18_SF_deeply_map network
Issue -
State: closed - Opened by daixiaolei623 over 3 years ago
- 1 comment
#42 - also question about fps
Issue -
State: closed - Opened by daixiaolei623 over 3 years ago
- 2 comments
#41 - transform:failed to synchronize: cudaErrorAssert: device-side assert triggered
Issue -
State: closed - Opened by Wu-ZJ over 3 years ago
- 2 comments
#40 - question about fps
Issue -
State: closed - Opened by songqi-github over 3 years ago
- 1 comment
#39 - FAM模块在反向传播的时候权重会更新吗
Issue -
State: closed - Opened by dd1ge almost 4 years ago
- 2 comments
#38 - Error: CUDA out of memory
Issue -
State: closed - Opened by daixiaolei623 almost 4 years ago
- 16 comments
Labels: bug
#37 - Unable to reproduce train
Issue -
State: closed - Opened by lucasjinreal almost 4 years ago
- 6 comments
Labels: bug
#37 - Unable to reproduce train
Issue -
State: closed - Opened by lucasjinreal almost 4 years ago
- 6 comments
Labels: bug
#36 - Unable to train
Issue -
State: closed - Opened by lucasjinreal almost 4 years ago
- 14 comments
Labels: bug
#35 - How did u train sfnet_resnet_dsn??
Issue -
State: closed - Opened by lucasjinreal almost 4 years ago
- 2 comments
#34 - about align_corners in grid_sample
Issue -
State: closed - Opened by lingorX almost 4 years ago
- 1 comment
#33 - Demo still invalid, Please provide pretrained weigth
Issue -
State: closed - Opened by lucasjinreal almost 4 years ago
- 6 comments
Labels: good first issue
#32 - 直接打印模型结构和提供的训练好的参数对应不太一致?
Issue -
State: closed - Opened by hanjr92 almost 4 years ago
- 4 comments
#31 - Pooling in inference mode
Issue -
State: closed - Opened by rd-wei almost 4 years ago
- 1 comment
#30 - problem of training train_cityscapes_sfnet_dfv1.sh
Issue -
State: closed - Opened by XiaoJiNu almost 4 years ago
- 1 comment
#29 - The TensorRT implementation
Issue -
State: closed - Opened by YLyeliang almost 4 years ago
- 1 comment
#28 - The implementation seems a little different from the paper
Issue -
State: closed - Opened by YLyeliang almost 4 years ago
- 1 comment
#27 - Can you provide the test code?
Issue -
State: closed - Opened by wangm9766 almost 4 years ago
- 1 comment
#26 - 用箭头可视化流场
Issue -
State: closed - Opened by ahhLS almost 4 years ago
- 1 comment
#25 - 请问是否可以提供deep-stem的ResNet50和ResNet101的预训练模型
Issue -
State: closed - Opened by ahhLS almost 4 years ago
- 1 comment
#24 - DFv1作为backbone训练
Issue -
State: closed - Opened by ahhLS almost 4 years ago
- 2 comments
#23 - The ResNet101 SFNet can't get the mIoU 81.2?
Issue -
State: closed - Opened by wangm9766 about 4 years ago
#22 - visualize result is weired
Issue -
State: closed - Opened by lucasjinreal about 4 years ago
- 11 comments
#21 - 关于FAM代码和torchcv里面有些许不同,??
Issue -
State: closed - Opened by swjtulinxi about 4 years ago
- 1 comment
#20 - Res50/Res101 预训练权重文件
Issue -
State: closed - Opened by ahhLS about 4 years ago
- 3 comments
#19 - Singel GPU with 8 memory
Issue -
State: closed - Opened by Lufei-github about 4 years ago
- 2 comments
Labels: question
#18 - Is the coarse data in Cityscapes dataset used for training ?
Issue -
State: closed - Opened by Lufei-github about 4 years ago
- 1 comment
#17 - 请问能否提供一个demo,选择可用模型对单张或者多张图片进行处理的代码例子?
Issue -
State: closed - Opened by jacob081 about 4 years ago
- 4 comments
Labels: enhancement
#16 - 请问如何用自己训练的模型在特定数据集上进行测试呢?
Issue -
State: closed - Opened by ahhLS about 4 years ago
- 2 comments
Labels: question
#15 - 关于Cityscapes的num_classes
Issue -
State: closed - Opened by ahhLS about 4 years ago
- 2 comments
#14 - 请问怎么可视化流场呢,文件夹下提供了一些可视化函数。但是不知道怎么开始可视化啊,请问能不能大概说下
Issue -
State: closed - Opened by swjtulinxi about 4 years ago
- 2 comments
#13 - 为什么这个代码复现精度要比论文低呢,,,还有代码可不可以可视化学到的流场啊,没看见这部分代码在哪里啊
Issue -
State: closed - Opened by swjtulinxi about 4 years ago
- 2 comments
#12 - Question about the result of sf-dfnet
Issue -
State: closed - Opened by Pedrojj97 about 4 years ago
- 3 comments
Labels: question
#11 - The pre-training model Resnet18-deep-Inplane128.pth is pre-trained in which data set? ImageNet or Mapillatus?
Issue -
State: closed - Opened by GitHubLiaoYong about 4 years ago
- 1 comment
Labels: question
#10 - 后续会发布速度测试模块么?
Issue -
State: closed - Opened by MichaelFan01 about 4 years ago
- 1 comment
Labels: enhancement
#9 - 后续会发布速度测试模块么,看论文里有说we re-implement grid sampling operator by CUDA to be used together with TensorRT
Issue -
State: closed - Opened by MichaelFan01 about 4 years ago
#8 - Using my own model to execute the eval. Py file on the validation set MIoU is only 2 points over.
Issue -
State: closed - Opened by GitHubLiaoYong about 4 years ago
- 3 comments
#7 - which parameter controls the sampling?
Issue -
State: closed - Opened by YongjieYang-YY about 4 years ago
- 2 comments
#6 - cityscapes_train_cv0_tile1024.json's purpose
Issue -
State: closed - Opened by chenjun2hao about 4 years ago
- 1 comment
Labels: question
#5 - mIoU is lower than 79%
Issue -
State: closed - Opened by mosjoker about 4 years ago
- 4 comments
Labels: bug, good first issue
#4 - 关于测试问题
Issue -
State: closed - Opened by hudfdfdf about 4 years ago
- 2 comments
#3 - when?wait in hope。
Issue -
State: closed - Opened by zhangyunming over 4 years ago
- 2 comments
#2 - Could you please explain why the performance of SFNet didn't reach the result claimed in the paper ?
Issue -
State: closed - Opened by seekFire over 4 years ago
- 4 comments
#1 - 关于torchcv的sfnet实现
Issue -
State: closed - Opened by stillwaterman over 4 years ago
- 3 comments