Mathematically, I don’t understand exactly what it means global_mean_pool in torch_geometric.nn.

My data is loaded by dataloader(batch_size=32)

like " x = global_mean_pool(x,batch)". In this case, a shape of x is (900,6) and the shape of batch is (900). As a result, the shape of x=global_mean_pool(x,batch) is (32,6).

what is mean… could anyone help me plz…