Performance issue in the definition of loss_ops_preexisting_no_penalty, kglib/kgcn/learn/loss.py(P1)

[DLPerf]

Hello, I found a performance issue in the definition of loss_ops_preexisting_no_penalty, kglib/kgcn/learn/loss.py, tf.math.reduce_any and tf.math.not_equal will be calculated repeatedly during program execution, resulting in reduced efficiency. I think they should be created before the loop.

The same issues exist in

• tf.boolean_mask(target_op.nodes, node_mask_op)
• tf.cast(features[:, 0], tf.int32)
• tf.expand_dims(feat_types, axis=1)

Looking forward to your reply. Btw, I am very glad to create a PR to fix it if you are too busy.

[jmsfltchr]

Looks like you're right! Yes please do submit PR(s) to fix the issues you see!