-
### Bug summary
When `rcut` is an int, an error throws: `Expected a value of type 'float' for argument 'rcut' but instead found type 'int'.`
### DeePMD-kit Version
b86165d2f2f49318c8d25fce8c6889bed…
njzjz updated
2 weeks ago
-
Running Jax on Apple silicon with the Metal backend (mps), might need quite some work.
Obviously, add "jax-metal" by running
`poetry add jax-metal`
which will add
`jax-metal = "^0.1.0"` to pypr…
-
### 🚀 The feature, motivation and pitch
Hi folks,
Thank you for your great effort in implementing KV cache compression methods in vLLM. I recently tried running experiments with tensor parallel …
-
### Feature Description
Instead of aggregating the output of segmentation attributions (using e.g. one-hot encoding) we can also aggregate the inputs for the spatial case.
### Use Case
This use cas…
-
hi
-
The Scaladoc should be similar to [numpy's API reference](https://docs.scipy.org/doc/numpy-1.13.0/reference/)
-
Hi,
I have implemented the forward pass using a convolution + sigmoid_fwd activation and am now working on the backpropagation of the graph. However, according to the [document](https://docs.nvidia.…
-
## 🚨🚨 Feature Request
- [ ] Related to an existing [Issue](../issues)
- [x] A new implementation (Improvement, Extension)
### If your feature will improve `HUB`
Currently, only Datasets can…
-
We already have `MaskedVisionTransformer` classes for TIMM and torchvision that take images as input and output a token for every image patch. These classes are versatile as they can be used for many …
-
A lot of code for tensor subclasses can likely be conslidated together into a base class that other classes can utilize
_get_to_kwargs:
https://github.com/pytorch/ao/blob/main/torchao/dtypes/affin…