-
The current backend architecture works well of out purpose. But it is not salable (in a programming sense) enough and will cause problems down the line. For the backend system to work, i.e. direct fun…
-
### 🚀 The feature, motivation and pitch
There currently exist NestedTensor which accepts a list of tensors but copies them into a new storage
There currently are a bunch of _foreach methods whic…
-
The Ollama model hub still has the default quant type of Q4_0 which is a legacy format that under-performs compared to K-quants (Qn_K, e.g. Q4_K_M, Q6_K, Q5_K_L etc...).
- Would it perhaps make sen…
-
Current `Module` and `ModuleT` traits can accept only Tensor type as input. It can be not suitable for complex models that should accept not only one tensor.
What about such trait for modules:
```ru…
-
For release 2.0, we plan to add support for multi-channel room impulse response simulation methods under `torchaudio.functional`. The implementation is based on [pyroomacoustics](https://github.com/LC…
-
Currently, the `full` function is implemented separately for each tensor type. To make it work for `bool` tensors as well, it was moved to the `BaseTensor` trait. However, this introduced some issues:…
-
The cameras [doc](https://pytorch3d.org/docs/cameras) says that:
1. In world coordinates the positive x is going left.
2. At screen space the top-left point is (0, 0) and the positive x goes to the …
-
**Describe the feature you'd like**
Now that we support `LocalSimulator`s, we can hook into [ITensor](https://github.com/ITensor/ITensors.jl)'s support for DMRG-based evolution methods to implement t…
-
**System information**
GPU : 3060, 12G
Linux Ubuntu 16.04
gcc6.3
bazel3.1
python3.8
cuda11.1
cudnn8.1
Tensorflow2.4.1
I use tensorflow by python api, which compiled by the source code, Got …
-
Hi, In **minitorch**, inappropriate dependency versioning constraints can cause risks.
Below are the dependencies and version constraints that the project is using
```
numpy==1.19.1
numba==0.4…