mindscience.e3nn

nn

`mindscience.e3nn.nn.Activation`	Activation function for scalar irreps (\(l = 0\)).
`mindscience.e3nn.nn.BatchNorm`	Batch normalization tailored for orthonormal group representations.
`mindscience.e3nn.nn.FullyConnectedNet`	Fully-connected Neural Network with normalized activation on scalars.
`mindscience.e3nn.nn.Gate`	Gate activation function.
`mindscience.e3nn.nn.NormActivation`	Activation function for the norm of irreps.
`mindscience.e3nn.nn.OneHot`	One-hot embedding with irreps support.
`mindscience.e3nn.nn.Scatter`	Easy-to-use wrapper for scatter operations: aggregates source values into a destination tensor according to index.
`mindscience.e3nn.nn.soft_one_hot_linspace`	Projection on a basis of functions.
`mindscience.e3nn.nn.soft_unit_step`	Smooth version of the unit step function.
`mindscience.e3nn.nn.SoftOneHotLinspace`	Projection on a basis of functions.

o3

`mindscience.e3nn.o3.angles_to_matrix`	Convert Euler angles (\(\alpha\), \(\beta\), \(\gamma\)) into the corresponding \(3 \times 3\) rotation matrix.
`mindscience.e3nn.o3.angles_to_xyz`	Convert the two spherical angles (\((\alpha, \beta)\)) into Cartesian coordinates \((x, y, z)\) on the unit sphere.
`mindscience.e3nn.o3.change_basis_real_to_complex`	Transform a real-valued spherical-harmonic basis into its complex-valued counterpart.
`mindscience.e3nn.o3.compose_angles`	Compute the Euler angles that result from composing two rotations.
`mindscience.e3nn.o3.ElementwiseTensorProduct`	Elementwise tensor product.
`mindscience.e3nn.o3.FullTensorProduct`	Full tensor product between two irreps.
`mindscience.e3nn.o3.FullyConnectedTensorProduct`	Fully-connected weighted tensor product.
`mindscience.e3nn.o3.identity_angles`	Return the identity set of Euler angles \((\alpha, \beta, \gamma)\) that corresponds to "no rotation".
`mindscience.e3nn.o3.Irrep`	Irreducible representation of O(3).
`mindscience.e3nn.o3.Irreps`	Direct sum of irreducible representations of O(3).
`mindscience.e3nn.o3.Linear`	Linear operation equivariant.
`mindscience.e3nn.o3.LinearBias`	Linear operation equivariant with option to add bias.
`mindscience.e3nn.o3.matrix_to_angles`	Convert \(3 \times 3\) rotation matrix into Euler angles (\((\alpha, \beta, \gamma)\)).
`mindscience.e3nn.o3.matrix_x`	Return the \(3 \times 3\) rotation matrix for a rotation about the x-axis by the given angle.
`mindscience.e3nn.o3.matrix_y`	Return the \(3 \times 3\) rotation matrix for a rotation about the y-axis by the given angle.
`mindscience.e3nn.o3.matrix_z`	Return the \(3 \times 3\) rotation matrix for a rotation about the z-axis by the given angle.
`mindscience.e3nn.o3.Norm`	Compute the norm (length) of each irreducible representation (irrep) contained in a direct-sum tensor.
`mindscience.e3nn.o3.rand_angles`	Return a uniformly-random set of Euler angles \((\alpha, \beta, \gamma)\) that represents a random rotation in 3-D space.
`mindscience.e3nn.o3.so3_generators`	Compute the so(3) Lie algebra generators.
`mindscience.e3nn.o3.spherical_harmonics`	Compute spherical harmonics.
`mindscience.e3nn.o3.SphericalHarmonics`	Spherical-harmonics cell: maps 3-D Cartesian vectors (x, y, z) to the corresponding complex-valued spherical-harmonic basis functions \(Y_l^m(\hat{x})\).
`mindscience.e3nn.o3.su2_generators`	Compute the su(2) Lie algebra generators.
`mindscience.e3nn.o3.TensorProduct`	Versatile tensor product operator of two input Irreps and a output Irreps, that sends two tensors into a tensor and keep the geometric tensor properties.
`mindscience.e3nn.o3.TensorSquare`	Compute the square tensor product of a tensor.
`mindscience.e3nn.o3.wigner_3j`	Wigner 3j symbols \(C_{lmn}\).
`mindscience.e3nn.o3.wigner_D`	Wigner D matrix representation of SO(3).
`mindscience.e3nn.o3.xyz_to_angles`	Convert a point \(\vec r = (x, y, z)\) on the sphere into angles \((\alpha, \beta)\).

so2_conv

`mindscience.e3nn.so2_conv.SO3Rotation`	Class for handling SO(3) rotations of spherical-harmonic irreps.
`mindscience.e3nn.so2_conv.SO2Convolution`	SO(2)-equivariant convolution layer for complex-valued features on the circle group.
`mindscience.e3nn.so2_conv.init_edge_rot_mat`	Initialize the rotation matrix from the edge distance vector.

utils

`mindscience.e3nn.utils.Ncon`	Multiple-tensor contraction operator which has similar function to Einsum.
`mindscience.e3nn.utils.radius`	Find all points in x for each element in y within distance r.
`mindscience.e3nn.utils.radius_full`	Find all points in x for each element in y.
`mindscience.e3nn.utils.radius_graph`	Computes graph edges to all points within a given distance.
`mindscience.e3nn.utils.radius_graph_full`	Computes graph edges to all points within a given distance.