hiplayers module

Full Documentation for hippynn.layers.hiplayers module. Click here for a summary page.

Layers for HIP-NN

class CosCutoff(hard_max_dist)

Bases: Module

forward(dist_tensor)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class GaussianSensitivityModule(n_dist, min_dist_soft, max_dist_soft, hard_max_dist, cutoff_type=<class 'hippynn.layers.hiplayers.CosCutoff'>)

Bases: SensitivityModule

forward(distflat, warn_low_distances=None)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class InteractLayer(nf_in, nf_out, n_dist, mind_soft, maxd_soft, hard_cutoff, sensitivity_module, cusp_reg=None)

Bases: Module

Hipnn’s interaction layer

forward(in_features, pair_first, pair_second, dist_pairs)

Pytorch Enforced Forward function

Parameters:

• in_features

• pair_first

• pair_second

• dist_pairs

Returns:

Interaction output features

regularization_params()

class InteractLayerQuad(nf_in, nf_out, n_dist, mind_soft, maxd_soft, hard_cutoff, sensitivity_module, cusp_reg)

Bases: InteractLayerVec

forward(in_features, pair_first, pair_second, dist_pairs, coord_pairs)

Pytorch Enforced Forward function

Parameters:

• in_features

• pair_first

• pair_second

• dist_pairs

Returns:

Interaction output features

class InteractLayerVec(nf_in, nf_out, n_dist, mind_soft, maxd_soft, hard_cutoff, sensitivity_module, cusp_reg)

Bases: InteractLayer

static compatibility_hook(self, incompatible_keys)

forward(in_features, pair_first, pair_second, dist_pairs, coord_pairs)

Pytorch Enforced Forward function

Parameters:

• in_features

• pair_first

• pair_second

• dist_pairs

Returns:

Interaction output features

get_extra_state()

Return any extra state to include in the module’s state_dict.

Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns:

object: Any extra state to store in the module’s state_dict

set_extra_state(state)

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Args:

state (dict): Extra state from the state_dict

class InverseSensitivityModule(n_dist, min_dist_soft, max_dist_soft, hard_max_dist, cutoff_type=<class 'hippynn.layers.hiplayers.CosCutoff'>)

Bases: SensitivityModule

forward(distflat, warn_low_distances=None)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class SensitivityBottleneck(n_dist, min_soft_dist, max_dist_soft, hard_max_dist, n_dist_bare, cutoff_type=<class 'hippynn.layers.hiplayers.CosCutoff'>, base_sense=<class 'hippynn.layers.hiplayers.InverseSensitivityModule'>)

Bases: Module

forward(distflat)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class SensitivityModule(hard_max_dist, cutoff_type)

Bases: Module

warn_if_under(distance, threshold)