Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
e3nn-mlx 0.1.0
e3nn-mlx 0.1.0
  • User guide
    • Installation
    • Irreducible representations
    • Tensor products
    • Testing equivariance
    • Migrating from e3nn/PyTorch
    • Performance and automatic differentiation
  • Examples
    • First equivariant operation
    • Equivariant graph convolution
    • Point-cloud models
  • API reference
    • O(3) operations
      • e3nn_mlx.o3.Irrep
      • e3nn_mlx.o3.MulIrrep
      • e3nn_mlx.o3.Irreps
      • e3nn_mlx.o3.Linear
      • e3nn_mlx.o3.Norm
      • e3nn_mlx.o3.TensorProduct
      • e3nn_mlx.o3.FullyConnectedTensorProduct
      • e3nn_mlx.o3.FullTensorProduct
      • e3nn_mlx.o3.ElementwiseTensorProduct
      • e3nn_mlx.o3.TensorSquare
      • e3nn_mlx.o3.ReducedTensorProducts
      • e3nn_mlx.o3.spherical_harmonics
      • e3nn_mlx.o3.spherical_harmonics_alpha_beta
      • e3nn_mlx.o3.SphericalHarmonics
      • e3nn_mlx.o3.SphericalHarmonicsAlphaBeta
      • e3nn_mlx.o3.s2_grid
      • e3nn_mlx.o3.ToS2Grid
      • e3nn_mlx.o3.FromS2Grid
      • e3nn_mlx.o3.SO3Grid
      • e3nn_mlx.o3.rand_angles
      • e3nn_mlx.o3.rand_matrix
      • e3nn_mlx.o3.rand_quaternion
      • e3nn_mlx.o3.rand_axis_angle
      • e3nn_mlx.o3.angles_to_matrix
      • e3nn_mlx.o3.matrix_to_angles
      • e3nn_mlx.o3.axis_angle_to_matrix
      • e3nn_mlx.o3.matrix_to_axis_angle
      • e3nn_mlx.o3.quaternion_to_matrix
      • e3nn_mlx.o3.matrix_to_quaternion
      • e3nn_mlx.o3.compose_angles
      • e3nn_mlx.o3.compose_quaternion
      • e3nn_mlx.o3.compose_axis_angle
      • e3nn_mlx.o3.wigner_d
      • e3nn_mlx.o3.irreps_wigner_d
      • e3nn_mlx.o3.irreps_wigner_d_from_matrix
      • e3nn_mlx.o3.wigner_3j
    • Neural-network modules
      • e3nn_mlx.nn.Activation
      • e3nn_mlx.nn.BatchNorm
      • e3nn_mlx.nn.Dropout
      • e3nn_mlx.nn.Extract
      • e3nn_mlx.nn.ExtractIr
      • e3nn_mlx.nn.FullyConnectedNet
      • e3nn_mlx.nn.Gate
      • e3nn_mlx.nn.Identity
      • e3nn_mlx.nn.NormActivation
      • e3nn_mlx.nn.S2Activation
      • e3nn_mlx.nn.SO3Activation
    • Mathematical helpers
      • e3nn_mlx.math.soft_unit_step
      • e3nn_mlx.math.soft_one_hot_linspace
      • e3nn_mlx.math.smooth_cutoff
    • Graph operations
      • e3nn_mlx.radius_graph
      • e3nn_mlx.scatter_sum
    • Models
      • e3nn_mlx.models.gate_points_2102.Convolution
      • e3nn_mlx.models.gate_points_2102.Network
      • e3nn_mlx.models.v2106.Convolution
      • e3nn_mlx.models.v2106.MessagePassing
      • e3nn_mlx.models.v2106.SimpleNetwork
      • e3nn_mlx.models.v2106.NetworkForAGraphWithAttributes
    • Representation-aware arrays
      • e3nn_mlx.IrrepsArray
  • Development notes
    • Compatibility and numerical conventions
    • e3nn-style high-level API
    • Upstream tests/o3 compatibility matrix
    • Upstream tests/nn compatibility matrix
    • gate_points_2102 MLX model
    • Modular v2106 point models
Back to top
View this page
Edit this page

ExamplesΒΆ

These examples emphasize complete computational patterns rather than individual signatures. They use ordinary MLX arrays and the e3nn-style API.

  • First equivariant operation
  • Equivariant graph convolution
  • Point-cloud models
Next
First equivariant operation
Previous
Performance and automatic differentiation
Copyright © 2026, lamalab-org
Made with Sphinx and @pradyunsg's Furo