Comments (5)
Hello,
The input_pts are the point coordinates of input point cloud => pos in your case. The format is [Batch size, Dim, NPoints]
The features are inputs_fts or x, depending on the example => y in your case.
from lightconvpoint.
in your model for classification task,
there is def forward(self, x, input_pts):
.
I write my train model like
`def train(epoch):
model.train()
for data in train_loader:
data = data.to(device)
optimizer.zero_grad()
loss = F.nll_loss(model(data.x, data.pos), data.y)
loss.backward()
optimizer.step()`
i simply injected your network for classification for my model, However, it did not work. i am wondering if there is no data.x in Data(pos=[10240, 3], y=[1]) .
What would you suggest a proper way to implement ConvPoint classification for your own dataset like I mentioned.
from lightconvpoint.
You need to reshape the data to fit input_pts.shape = [BatchSize, Dim, NPoints] and x.shape = [BatchSize, C, NPoints]
For example with a batchsize=1 and x=1 you can do:
npoints = data.pos.shape[1]
x = torch.ones(1, 1, npoints ).to(device)
input_pts = data.pos.transpose(0,1).unsqueeze(0)
then:
outputs = model(x, input_pts)
from lightconvpoint.
(https://github.com/rusty1s/pytorch_geometric/blob/master/examples/pointnet2_classification.py))
Isnt his example already reshape the data?
`
i am really new to this.
could you give me more specific instruction for reshaping data in which steps? Loading data in trainloader, Forward pass, trainning loop?
from lightconvpoint.
Hello,
I am not familiar with pytorch_geometric. I do not about the data format.
from lightconvpoint.
Related Issues (15)
- I am very interested in your work, but A problem accured when install the setup.py. Do you kown why? HOT 1
- No 'libtorch_cpu.so' for knn
- Install ERROR: Command errored out with exit status 1 HOT 9
- Install Error HOT 1
- error about installing package in project HOT 1
- Space quantization
- question
- time consuming in test.py
- SemanticKITTI benchmark HOT 1
- NPM3D Benchmark HOT 2
- If LightConvPoint is faster than Convpoint for training Modelnet40
- Is KPconv kinda implemented in this repo? HOT 3
- Setup.py HOT 4
- validation
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 lightconvpoint.