Asad Abbas

Mixed precision (16-bit) training error

7

I'm trying to use 16 bit precision in pytorch lightning to save some gpu memory, but I'm getting this error: `RuntimeError:MinkowskiEngine/src/convolution_gpu.cu:69, assertion (in_feat.scalar_type() == kernel.scalar_type()) failed. type mismatch` Is there...

Thanks, @piEsposito for the amazing library. I'm using a couple of Bayesian linear layers at the end of a model for regression analysis. One thing that I noticed is that...

Thanks for the awesome work. I'm trying to use RandLA for a regression task. I'm actually using the PyTorch implementation and tried replacing the final layers with linear layers but...

Trained Model

2

Thanks for your work. Unfortunately, I don't have sufficient resources to train model myself. Do you plan to release trained checkpoint model anytime soon? It would be good to provide...

Thanks, @JonasSchult for sharing your work. I'm trying to prepare a dataset for a regression task. I'm not sure how to handle crops as I think the following piece of...

Incorporating mesh level properties

1

Thanks for sharing your awesome work. I'm trying to work with inputs that have the following shapes: mesh coordinates: x, y, z mesh level scalar parameters: e.g: `A=10, B=12, C=150,...

Working with meshes having different number of vertices?

1

Thanks for sharing your code. I have a dataset of 3D meshes having a different number of vertices. The size of convolution layers depends on indices parameter and they are...

Cleaned ShapeNet and simulated pressure Dataset

1

Thanks for awesome work. As mentioned in the paper, do you guys still have plans to make both the cleaned car split of ShapeNet and the simulated pressure values publicly...

PatchTST with y shape of (n_samples, n_outputs)

5

Thanks for implementing PatchTST :) I'm trying to follow the tutorial provided for PatchTST. The documentation of TSForecaster says: ``` X: array-like of shape (n_samples, n_steps) or (n_samples, n_features, n_steps)...

Thanks for sharing the test code. Do you plan to share the training code? I would like to train it on my own custom dataset.