Implement more pooling operators

[CarloLucibello]

This is the list of pooling operators in pytorch geometric

• [x] global_add_pool (GlobalPool(+) here)
• [x] global_mean_pool (GlobalPool(mean) here)
• [x] global_max_pool (GlobalPool(max) here)
• [ ] global_sort_pool
• [x] GlobalAttention (GlobaAttentionlPool(max) here)
• [ ] Set2Set
• [ ] GraphMultisetTransformer

[zsz00]

what's the status of this now?

[CarloLucibello]

I'm not actively working on pooling operators right now

[zsz00]

I read doc of pool: GraphNeuralNetworks.GlobalAttentionPool GraphNeuralNetworks.GlobalPool GraphNeuralNetworks.TopKPool

Doesn't that already exist?

[CarloLucibello]

I updated the list with the current status. TopK I think is broken and has to be completely reworked. Also, we don't have sort pooling and set2set.