Open AndreaPi opened 3 days ago
@Alexey-Kamenev could you please take a look?
You are correct, BSMS MGN expects the data in a certain format. To enable this format, you need to wrap your dataset class in BistrideMultiLayerGraphDataset
like it's done in the Ahmed body example. You can do this either in the code or by using Hydra config - check out the BSMS Ahmed body experiment and corresponding dataset config.
Version
0.8.0
On which installation method(s) does this occur?
Docker
Describe the issue
I'm trying to train a BiStride MeshGraphNet on my dataset (very similar to DrivAerNet), but I keep getting errors. It looks like it's expecting the data in the graph to have a very specific structure, unlike MeshGraphNet which is better written (and it trains on my data). The error I'm getting is
Can you help? It would be useful if you provided an example to test BiStride MeshGraphNet out, but the only example mentioned in the documentation regards the Ahmed body dataset which is not included in the
examples
folder. https://docs.nvidia.com/deeplearning/modulus/modulus-core/examples/cfd/aero_graph_net/readme.html#bsms-mgn-trainingMinimum reproducible example
This is the dataset class:
And this is the
__init__
method of my trainer class:Relevant log output
No response
Environment details
No response