Comments (4)
This happened because the point locations may be out of range. I add the clip()
function to avoid it, as shown in line-56:
points[:, 1] = torch.clip(points[:, 1], min=0, max=h-1)
points[:, 0] = torch.clip(points[:, 0], min=0, max=w-1)
from spdcn-cac.
Thanks for your thoughtful reply, the problem has been solved.
By the way, dose the code you provide allow for multi-GPU training?
from spdcn-cac.
I'm afraid the answer is no.
However, the batch size is 16, and the memory requirement is about 21G, so one 3090-GPU is enough to train this model. You can reduce the batch size and adjust the learning rate accordingly for training with limited resources.
Another way to reduce computation costs is using FP16, which may decrease the accuracy a little but save lots of memory.
from spdcn-cac.
Thanks for your thoughtful reply.
from spdcn-cac.
Related Issues (6)
- Inference HOT 3
- mmcv issue HOT 1
- How can i use HOT 4
- demo.ipynb HOT 4
- Program syntax error 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 spdcn-cac.