Comments (7)
Could you provide the detailed config and running command?
from sssegmentation.
您好,我想复现的是基于backbone是resnet50条件下,mcibi++在Cityscapes数据集上的结果,在配置文件为configs/base_config.py中第84行将context_within_image的is_on设置为false,并且在configs/memorynetv2/memorynetv2_resnet50os8_cityscapes.py中也没有配置is_on为True,但我发现其他的配置文件中将is_on重新设置为True,可以帮我解答一下嘛
from sssegmentation.
您好,我还有个问题,我没有找到FCN+MCIBI++的配置文件,好像只有ASPP,PPM,UPERNET+MCIBI++的配置文件
from sssegmentation.
https://github.com/SegmentationBLWX/sssegmentation/tree/main/docs/performances/memorynetv2 table里都有对应的config路径
from sssegmentation.
您好,我想复现的是基于backbone是resnet50条件下,mcibi++在Cityscapes数据集上的结果,在配置文件为configs/base_config.py中第84行将context_within_image的is_on设置为false,并且在configs/memorynetv2/memorynetv2_resnet50os8_cityscapes.py中也没有配置is_on为True,但我发现其他的配置文件中将is_on重新设置为True,可以帮我解答一下嘛
你运行命令是?
from sssegmentation.
不好意思,之前我看错了,我还有个问题,就是你代码的modules/models/segmentors/memorynet/memory.py中的第214行代码,这个relation和argmax对应的变量是什么意思呀,可以为我解答一下嘛,谢谢您。
assert strategy in ['cosine_similarity'] # ----(K, C) * (C, num_feats_per_cls) --> (K, num_feats_per_cls) relation = torch.matmul( F.normalize(feats_cls, p=2, dim=1), F.normalize(self.memory[clsid].data.permute(1, 0).contiguous(), p=2, dim=0), ) argmax = relation.argmax(dim=1)# 返回最大值的索引 # ----for saving memory during training for idx in range(self.num_feats_per_cls): mask = (argmax == idx) feats_cls_iter = feats_cls[mask] memory_cls_iter = self.memory[clsid].data[idx].unsqueeze(0).expand_as(feats_cls_iter) similarity = F.cosine_similarity(feats_cls_iter, memory_cls_iter) weight = (1 - similarity) / (1 - similarity).sum() feats_cls_iter = (feats_cls_iter * weight.unsqueeze(-1)).sum(0) self.memory[clsid].data[idx].copy_(self.memory[clsid].data[idx] * (1 - momentum) + feats_cls_iter * momentum)
from sssegmentation.
您好,感谢你卓越的贡献,我想请问关于MCIBI++代码的context_within_image_cfg的is_on设置为false,跟论文里面的不一致,请问可以帮我解答一下嘛,谢谢
您好,我也在学习MCIBI++,但是我之前并未使用过sssegmentation,在工具箱运行方面遇到了一些问题,想跟您请教一下,也方便之后一起探讨和进步,如果您有时间的话,可以加一下我的微信:ankeas,万分感谢!!!!!我目前在北京理工大学学习
from sssegmentation.
Related Issues (20)
- ssseg/modules/models/segmentors/memorynet/initmemory.py HOT 3
- 模型复杂度计算和FLOPS HOT 2
- 关于每个数据集的通道正则化 HOT 7
- 关于测试
- Where is your difference between mmseg HOT 1
- 是否可以对在Windows环境下的单卡环境进行一定的优化呢,现在的整体代码都是多卡环境用起来很卡手
- 关于VSPW数据集训练 HOT 1
- Do we have any usage examples? HOT 2
- [BUG]: ERROR: Failed building wheel for chainercv HOT 1
- [BUG] HOT 1
- [BUG] HOT 1
- [note: This error originates from a subprocess, and is likely not a problem with pip. ERROR: Failed building wheel for chainercv Running setup.py clean for chainercv Failed to build chainercv ERROR: Could not build wheels for chainercv, which is required to install pyproject.toml-based projects] HOT 1
- IDRNet HOT 1
- ERROR: Could not build wheels for chainercv, which is required to install pyproject.toml-based projects HOT 17
- errrors recorded while attemplting the training, (remember that i removed the chainercv from the requirements)
- torchtext 0.8.1 requires torch==1.7.1 HOT 1
- error during training HOT 24
- assert num_classes == len(classnames) and num_classes == len(palette) HOT 6
- 关于单机单卡条件下训练模型
- 训练刚结束时的验证效果和保存并重新加载模型参数后进行验证的结果不一样 HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from sssegmentation.