torch-harmonics implements differentiable signal processing on the sphere. This includes differentiable implementations of the spherical harmonic transforms, vector spherical harmonic transforms and discrete-continuous convolutions on the sphere. The package was originally implemented to enable Spherical Fourier Neural Operators (SFNO) [1].
The SHT algorithm uses quadrature rules to compute the projection onto the associated Legendre polynomials and FFTs for the projection onto the harmonic basis. This algorithm tends to outperform others with better asymptotic scaling for most practical purposes [2].
torch-harmonics uses PyTorch primitives to implement these operations, making it fully differentiable. Moreover, the quadrature can be distributed onto multiple ranks making it spatially distributed.
torch-harmonics has been used to implement a variety of differentiable PDE solvers which generated the animations below. Moreover, it has enabled the development of Spherical Fourier Neural Operators [1].
A simple installation can be directly done from PyPI:
```bash
pip install torch-harmonics
```
If you are planning to use spherical convolutions, we recommend building the corresponding custom CUDA kernels. To enforce this, you can set the `FORCE_CUDA_EXTENSION` flag. You may also want to set appropriate architectures with the `TORCH_CUDA_ARCH_LIST` flag. Finally, make sure to disable build isolation via the `--no-build-isolation` flag to ensure that the custom kernels are built with the existing torch installation.
docker run --gpus all -it--rm--ipc=host --ulimitmemlock=-1--ulimitstack=67108864 torch_harmonics
```
```
## More about torch-harmonics
### 2、使用源码编译方式安装
### Spherical harmonics
The [spherical harmonics](https://en.wikipedia.org/wiki/Spherical_harmonics) are special functions defined on the two-dimensional sphere $S^2$ (embedded in three dimensions). They form an orthonormal basis of the space of square-integrable functions defined on the sphere $L^2(S^2)$ and are comparable to the harmonic functions defined on a circle/torus. The spherical harmonics are defined as
where $\theta$ and $\lambda$ are colatitude and longitude respectively, and $P_l^m$ the normalized, [associated Legendre polynomials](https://en.wikipedia.org/wiki/Associated_Legendre_polynomials).
realizes the projection of a signal $f(\theta, \lambda)$ on $S^2$ onto the spherical harmonics basis. The SHT generalizes the Fourier transform on the sphere. Conversely, a truncated series expansion of a function $f$ can be written in terms of spherical harmonics as
where $\hat{f}_l^m$, are the expansion coefficients associated to the mode $m$, $n$.
The implementation of the SHT follows the algorithm as presented in [2]. A direct spherical harmonic transform can be accomplished by a Fourier transform
The second integral, which computed the projection onto the Legendre polynomials is realized with quadrature. On the Gaussian grid, we use Gaussian quadrature in the $\cos \theta$ domain. The integral
Here, $x_j \in [-1,1]$ are the quadrature nodes with the respective quadrature weights $w_j$.
### Discrete-continuous convolutions on the sphere
torch-harmonics now provides local discrete-continuous (DISCO) convolutions as outlined in [5] on the sphere. These are use in local neural operators [2] to generalize convolutions to structured and unstructured meshes on the sphere.
### Spherical (neighborhood) attention
torch-harmonics introducers spherical attention mechanisms which correctly generalize the attention mechanism to the sphere. The use of quadrature rules makes the resulting operations approximately equivariant and equivariant in the continuous limit. Moreover, neighborhood attention is correctly generalized onto the sphere by using the geodesic distance to determine the size of the neighborhood.
## Getting started
The main functionality of `torch_harmonics` is provided in the form of `torch.nn.Modules` for composability. A minimum example is given by:
9.[Resampling signals on the sphere](./notebooks/resample_sphere.ipynb)
## Examples and reproducibility
The `examples` folder contains training scripts for three distinct tasks:
*[solution of the shallow water equations on the rotating sphere](./examples/shallow_water_equations/train.py)
*[depth estimation on the sphere](./examples/depth/train.py)
*[semantic segmentation on the sphere](./examples/segmentation/train.py)
Results from the papers can generally be reproduced by running `python train.py`. In the case of some older results the number of epochs and learning-rate may need to be adjusted by passing the corresponding command line argument.
## Remarks on automatic mixed precision (AMP) support
Note that torch-harmonics uses Fourier transforms from `torch.fft` which in turn uses kernels from the optimized `cuFFT` library. This library supports fourier transforms of `float32` and `float64` (i.e. `single` and `double` precision) tensors for all input sizes. For `float16` (i.e. `half` precision) and `bfloat16` inputs however, the dimensions which are transformed are restricted to powers of two. Since data is converted to one of these reduced precision floating point formats when `torch.autocast` is used, torch-harmonics will issue an error when the input shapes are not powers of two. For these cases, we recommend disabling autocast for the harmonics transform specifically:
torch-harmonics implements differentiable signal processing on the sphere. This includes differentiable implementations of the spherical harmonic transforms, vector spherical harmonic transforms and discrete-continuous convolutions on the sphere. The package was originally implemented to enable Spherical Fourier Neural Operators (SFNO) [1].
The SHT algorithm uses quadrature rules to compute the projection onto the associated Legendre polynomials and FFTs for the projection onto the harmonic basis. This algorithm tends to outperform others with better asymptotic scaling for most practical purposes [2].
torch-harmonics uses PyTorch primitives to implement these operations, making it fully differentiable. Moreover, the quadrature can be distributed onto multiple ranks making it spatially distributed.
torch-harmonics has been used to implement a variety of differentiable PDE solvers which generated the animations below. Moreover, it has enabled the development of Spherical Fourier Neural Operators [1].
A simple installation can be directly done from PyPI:
```bash
pip install torch-harmonics
```
If you are planning to use spherical convolutions, we recommend building the corresponding custom CUDA kernels. To enforce this, you can set the `FORCE_CUDA_EXTENSION` flag. You may also want to set appropriate architectures with the `TORCH_CUDA_ARCH_LIST` flag. Finally, make sure to disable build isolation via the `--no-build-isolation` flag to ensure that the custom kernels are built with the existing torch installation.
docker run --gpus all -it--rm--ipc=host --ulimitmemlock=-1--ulimitstack=67108864 torch_harmonics
```
## More about torch-harmonics
### Spherical harmonics
The [spherical harmonics](https://en.wikipedia.org/wiki/Spherical_harmonics) are special functions defined on the two-dimensional sphere $S^2$ (embedded in three dimensions). They form an orthonormal basis of the space of square-integrable functions defined on the sphere $L^2(S^2)$ and are comparable to the harmonic functions defined on a circle/torus. The spherical harmonics are defined as
where $\theta$ and $\lambda$ are colatitude and longitude respectively, and $P_l^m$ the normalized, [associated Legendre polynomials](https://en.wikipedia.org/wiki/Associated_Legendre_polynomials).
realizes the projection of a signal $f(\theta, \lambda)$ on $S^2$ onto the spherical harmonics basis. The SHT generalizes the Fourier transform on the sphere. Conversely, a truncated series expansion of a function $f$ can be written in terms of spherical harmonics as
where $\hat{f}_l^m$, are the expansion coefficients associated to the mode $m$, $n$.
The implementation of the SHT follows the algorithm as presented in [2]. A direct spherical harmonic transform can be accomplished by a Fourier transform
The second integral, which computed the projection onto the Legendre polynomials is realized with quadrature. On the Gaussian grid, we use Gaussian quadrature in the $\cos \theta$ domain. The integral
Here, $x_j \in [-1,1]$ are the quadrature nodes with the respective quadrature weights $w_j$.
### Discrete-continuous convolutions on the sphere
torch-harmonics now provides local discrete-continuous (DISCO) convolutions as outlined in [5] on the sphere. These are use in local neural operators [2] to generalize convolutions to structured and unstructured meshes on the sphere.
### Spherical (neighborhood) attention
torch-harmonics introducers spherical attention mechanisms which correctly generalize the attention mechanism to the sphere. The use of quadrature rules makes the resulting operations approximately equivariant and equivariant in the continuous limit. Moreover, neighborhood attention is correctly generalized onto the sphere by using the geodesic distance to determine the size of the neighborhood.
## Getting started
The main functionality of `torch_harmonics` is provided in the form of `torch.nn.Modules` for composability. A minimum example is given by:
9.[Resampling signals on the sphere](./notebooks/resample_sphere.ipynb)
## Examples and reproducibility
The `examples` folder contains training scripts for three distinct tasks:
*[solution of the shallow water equations on the rotating sphere](./examples/shallow_water_equations/train.py)
*[depth estimation on the sphere](./examples/depth/train.py)
*[semantic segmentation on the sphere](./examples/segmentation/train.py)
Results from the papers can generally be reproduced by running `python train.py`. In the case of some older results the number of epochs and learning-rate may need to be adjusted by passing the corresponding command line argument.
## Remarks on automatic mixed precision (AMP) support
Note that torch-harmonics uses Fourier transforms from `torch.fft` which in turn uses kernels from the optimized `cuFFT` library. This library supports fourier transforms of `float32` and `float64` (i.e. `single` and `double` precision) tensors for all input sizes. For `float16` (i.e. `half` precision) and `bfloat16` inputs however, the dimensions which are transformed are restricted to powers of two. Since data is converted to one of these reduced precision floating point formats when `torch.autocast` is used, torch-harmonics will issue an error when the input shapes are not powers of two. For these cases, we recommend disabling autocast for the harmonics transform specifically:
