Batch operation

[lonelygoatherd]

Hi, thank for sharing the codes. In the networks.py, what does the variable "data.batch" denote in the forward method? I have looked up to pyg document, and found neither Data nor DataLoader has "batch" attribute. So what is the difference of it with the input x? 你好，请问networks里的forward函数，x和batch有什么不同呢，data有batch属性吗？找了半天没找到，但是运行都可以

[icecat2012]

I think the 'batch' variable is for the global_max_pool and global_mean_pool functions (readout). x1 = torch.cat([gmp(x, batch), gap(x, batch)], dim=1)

see pytorch geometric doc

batch (LongTensor) – Batch vector, which assigns each node to a specific example.