mindchemistry.e3.nn.soft_one_hot_linspace

mindchemistry.e3.nn.soft_one_hot_linspace(x, start, end, number, basis='smooth_finite', cutoff=True)[source]

Projection on a basis of functions. Returns a set of \(\{y_i(x)\}_{i=1}^N\),

\[y_i(x) = \frac{1}{Z} f_i(x)\]

where \(x\) is the input and \(f_i\) is the ith basis function. \(Z\) is a constant defined (if possible) such that,

\[\langle \sum_{i=1}^N y_i(x)^2 \rangle_x \approx 1\]

Note that bessel basis cannot be normalized.

Parameters

x (Tensor) – The shape of Tensor is \((...)\).
start (float) – minimum value span by the basis.
end (float) – maximum value span by the basis.
number (int) – number of basis functions \(N\).
basis (str) – {'gaussian', 'cosine', 'smooth_finite', 'fourier', 'bessel'}, the basis family. Default: 'smooth_finite'.
cutoff (bool) – whether require the \(y_i(x)\) from the outside domain of (start, end) to be vanished. Default: True.

Returns

Tensor, shape is \((..., N)\).

Raises

ValueError – If basis is not in {'gaussian', 'cosine', 'smooth_finite', 'fourier', 'bessel'}.

Supported Platforms:: Ascend

Examples

>>> from mindchemistry.e3.nn import soft_one_hot_linspace
>>> from mindspore import ops, Tensor
>>> x = Tensor(ops.ones((4, 6)))
>>> outputs = soft_one_hot_linspace(x, -0.5, 1.5, number=4)
>>> print(outputs.shape)
(4, 6, 4)