-
Hi, team! Thanks for your excellent work, although it is not open source.
I have a question about the “Convolutions on the Spatio-Temporal Graph” stage:
The message passing through **$G^{time}$** …
-
Hello!
Thank you for the repo. The paper mentions that the private encoders apply a position-wise MLP
with parameter to encode input X. Also, M independent temporal convolutions are applied with va…
-
https://arxiv.org/abs/1707.03141
Would love to see an implementation contributed to DeepChem. Wonder how it would compare to IterRef LSTMs
-
1.HoloPose: Holistic 3D Human Reconstruction In-The-Wild(2019)
mixture-of-experts rotation prior, part-based modeling( features co-varies with joint position)
code: can not open [http://arielai.com/…
-
Like the MLP autoregressive model but using temporal convolutions proposed in Temporal Convolutional Networks (TCN).
-
- [x] Dataset
- [ ] Basic visualization (follow #7 )
- [ ] Literature
- [x] #8
- [x] @Tanvig Add the other paper for which you found code here
- [ ] Implementing the (2+1)D model (used i…
-
I was reading the paper `Long-term Temporal Convolutions for Action Recognition` and read that they have tried different temporal extent `t ∈{20,40,60,80,100}` on the 60f Network.
I didn't get the…
-
As we expand OpenPIV-c--qt, I would like to add image filters. These filters would include the following:
- [x] 1D convolution filter (applies 1D convolution to each axis of the image instance)
-…
-
Does onnxrt enable `cudnn.benchmark = True` (in PyTorch lingo)? (I found https://github.com/microsoft/onnxruntime/pull/712 which suggests that benchmarking is done)
We're observing that onnxrt-gpu …
-
CuDNN docs mention that for optimal matmul (i.e. GEMM) performance, the input matrix sizes should be divisisible by 16 (or even 256 for best tiling):
- https://docs.nvidia.com/deeplearning/performanc…