magnet.aggmodels.refiner.Reyyy#

class magnet.aggmodels.refiner.Reyyy(n_features: int = 6, units: int = 10)#

Bases: DRLRefiner

Deep Reinforcement Learning refiner variant.

Graph Neural network with 2 SAGE convolutional layers followed by.

Constructor

__init__(n_features: int = 6, units: int = 10)#

Initialize internal Module state, shared by both nn.Module and ScriptModule.

Methods

__init__([n_features, units])

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(graph)

Define the computation performed at every call.
__init__(n_features: int = 6, units: int = 10)#

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(graph)#

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.

Inherited Methods

A2C_train(training_dataset[, ...])

agglomerate(mesh[, mode, nref, mult_factor])

Agglomerate a mesh.

agglomerate_dataset(dataset, **kwargs)

Agglomerate all meshes in a dataset.

bisect(mesh)

Bisect the mesh once.

bisection_Nref(mesh, Nref[, warm_start])

Bisect the mesh recursively a set number of times.

bisection_mult_factor(mesh, mult_factor[, ...])

Bisect a mesh until the agglomerated elements are small enough.

bisection_segregated(mesh, mult_factor[, subset])

Bisect heterogeneous mesh until elements are small enough.

change_vert(graph, action)

In place change of vertex to other subgraph.

coarsen(mesh, subset[, mode, nref, mult_factor])

Coarsen a subregion of the mesh.

compute_episode_length(graph)

cut(graph)

get_number_of_parameters()

Get total number of parameters of the GNN.

get_sample(mesh[, k, partitioner])

create a graph data structure sample from a mesh.

k_hop_graph_cut(graph, k)

Exrtact k-hop subgraph around the current cut fo refinement.

load_model(model_path)

Load model from state dictionary.

loss_function(output, graph)

Loss function used during training.

multilevel_bisection(mesh[, refiner, ...])

normalize(x)

Normalize the data before feeding it to the GNN.

objective(graph, starter)

reward_function(new_state, old_state, ...[, ...])

Modified normalized cut to take into account cell volumes instead

save_model(output_path)

Save current model to state dictionary.

train_GNN(training_dataset, ...[, ...])

Train the Graph Neural Network.

update_state(graph, action, nnz)

volumes(graph)