Propagational MLP (PMLP) on Cora
Author: K3-Node Team
Backend: Multi-Backend
Dataset: Cora (Planetoid)
Description: Classify papers in the Cora citation network by topic.
Propagational MLP (PMLP) on Cora
Classify papers in the Cora citation network by topic. A PMLP (Yang et al., 2023) is trained as a plain MLP that ignores the graph, and only uses message passing when making predictions. Surprisingly, this works about as well as a GCN.
Same model as PyG's examples/pmlp.py.
Install K3-Node, then choose a backend: "tensorflow", "torch" or "jax"
Load the data
data holds one graph: node features x, edges edge_index, labels y, and masks marking the training, validation and test nodes.
import keras
from keras import ops
from k3_node.datasets import Planetoid
from k3_node.loader import FullGraphDataset
from k3_node.transforms import NormalizeFeatures
dataset = Planetoid("data/Planetoid", name="Cora", transform=NormalizeFeatures())
data = dataset[0]
print(data)
Define the model
from k3_node.models import PMLP
class PMLPClassifier(keras.Model):
def __init__(self, in_channels, out_channels):
super().__init__()
self.pmlp = PMLP(in_channels, hidden_channels=16, out_channels=out_channels,
num_layers=2, dropout=0.5, norm=False)
def call(self, data, training=False):
# Train without the graph (like an MLP); use message passing for predictions
edge_index = None if training else data.edge_index
return self.pmlp(data.x, edge_index, training=training)
model = PMLPClassifier(dataset.num_features, dataset.num_classes)
Train
FullGraphDataset feeds the whole graph to Keras; mask selects which nodes count in the loss and the accuracy.
model.compile(
optimizer=keras.optimizers.Adam(learning_rate=0.01, weight_decay=0.0005),
loss=keras.losses.SparseCategoricalCrossentropy(from_logits=True),
weighted_metrics=["accuracy"],
)
model.fit(
FullGraphDataset(data, mask="train_mask"),
validation_data=FullGraphDataset(data, mask="val_mask"),
epochs=200,
verbose=2,
)