tensorclouds.nn¶
Submodules¶
- tensorclouds.nn.attention
- tensorclouds.nn.egnn
- tensorclouds.nn.embed
- tensorclouds.nn.feed_forward
- tensorclouds.nn.layer_norm
- tensorclouds.nn.pooling
- tensorclouds.nn.radius_graph
- tensorclouds.nn.residual
- tensorclouds.nn.self_interaction
- tensorclouds.nn.sequence
- tensorclouds.nn.spatial
- tensorclouds.nn.tensor_fields
- tensorclouds.nn.transformer
- tensorclouds.nn.utils
Classes¶
A TensorCloud is a collection of tensors with associated coordinates and masks. |
Package Contents¶
- class tensorclouds.nn.TensorCloud¶
A TensorCloud is a collection of tensors with associated coordinates and masks. It is used to represent a cloud of tensors in a 3D space, where each tensor can have different irreducible representations (irreps).
- irreps_array¶
The array of tensors with their irreducible representations.
- Type:
e3nn.IrrepsArray
- mask_irreps_array¶
A boolean mask indicating which irreps are present.
- Type:
jax.Array
- coord¶
The coordinates of the tensors in 3D space.
- Type:
jax.Array
- mask_coord¶
A boolean mask indicating which coordinates are valid.
- Type:
jax.Array
- label¶
An optional label for the TensorCloud, e.g., for classification tasks.
- Type:
jax.Array, optional
- irreps_array: e3nn_jax.IrrepsArray¶
- mask_irreps_array: jax.Array¶
- coord: jax.Array¶
- mask_coord: jax.Array¶
- label: jax.Array = None¶
- property shape¶
- classmethod empty(irreps: e3nn_jax.Irreps)¶
- classmethod ones_like(tc)¶
- classmethod zeros_like(tc)¶
- property irreps: e3nn_jax.Irreps¶
- property mask: jax.Array¶
- replace(**kwargs)¶
- classmethod concatenate(clouds: List[TensorCloud]) TensorCloud ¶
- classmethod zeros(irreps: e3nn_jax.Irreps, shape: tuple) TensorCloud ¶
- classmethod stack(clouds: List[TensorCloud]) TensorCloud ¶
- split(piece_sizes: List[int]) List[TensorCloud] ¶
- centralize()¶
- plot(view=None, viewer=None, colors=None, radius=0.04, mid=0.95)¶
- dot(other)¶
- norm()¶
Compute the norm of the irreps_array and coord of the TensorCloud. :returns: A tuple containing the norms of the irreps_array and coord. :rtype: Tuple[jax.Array, jax.Array]