Comments (7)
根据您的描述我无法确定问题出在哪里,您可以提供更多的信息,尤其是对代码的修改。
from step.
我没有对代码做任何的修改。。。所以我担心是预训练参数的差别。如果你们也没有对预训练参数做更多的调整,那么可能是机器的差异吧?
from step.
有点奇怪,我跑的代码和结果都在这个repo里面了,你也可以在training_logs看到我的结果。晚点我再跑一次试试吧。
from step.
您复现的结果是什么?不同的机器确实会有差异,但不会差很多才对。
from step.
在我完全复现pre-training+graph wavenet finetuning的结果中,虽然比论文中低一点,但也是SOTA。。
只是pre-training中的过程中,validation和test的指标跟github中的logs差距比较大,我可以把logs发到你在论文中留的邮箱中,因为logs太长了在这里不好post。
from step.
按理来说,预训练阶段的Train和Valid应该是一样的。Test确实会和我的log里面不一样,因为我后面修复过一个bug。See issue #18 . 预训练阶段的Test没有啥意义,可以不用关注。只要Downstream中表现正常就行。
STEP曾重构过,重构之后的结果在training_logs文件夹中,可能会和论文中的值有偏差。
METR-LA和PEMS-BAY会更好一点,PEMS04可能会差一点。但应该都还是SOTA,不会影响文章结论。您可以引用repo中的值而不用论文中的值。
from step.
我看了引用的issue,这确实解答了我的问题。您们的开源做的很好,感谢!
from step.
Related Issues (20)
- About STEP-DCRNN HOT 4
- 关于adj_mx.pkl这个数据集的问题 HOT 1
- 数据集替换 HOT 1
- Error HOT 6
- torch.cuda.OutOfMemoryError
- 作者你好,请问这个可以用单步预测里面吗,类似mtgnn 里exchage那种的
- 复现出现张量维度不匹配 HOT 3
- 作者您好,向您请教一下discrete_graph_learning.py中实现Equation(2)的细节
- launch_training() got an unexpected keyword argument 'devices'
- 请问如何用TSformer预训练模型,是否有readme文件介绍步骤 HOT 2
- 用自己的数据集来 预训练 模型时,如何配置 HOT 2
- 配置“CFG.RE_SCALE=False” HOT 3
- 单变量是否可以用这个预训练模型 HOT 2
- 数据集只有两列,请问是否可以用tsformer预训练模型 HOT 4
- About Figure 1 in the Paper HOT 4
- Setting of input and output length of compared model
- TSFormer预训练模型是否可以处理时间序列多通道数据? HOT 3
- 关于discrere_graph_learning.py 中的train_length HOT 1
- 多卡训练失败问题 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 step.