Kaká
Kaká
what you mean by sample amount? is it a train test split? the mean of token number is 231
I have 60k data and divided each train and test split by half equally, so train.txt and test.txt have 30k each
@AlanSwift btw, can you give me an example code to preprocess trec dataset so that the trec dataset have all the .pt files?
So, i just need to truncated the sentences?
Hey, I face the same error, after change `cfg.data.workers_per_gpu=0` there is another error occurs, it says: **ValueError: persistent_workers option needs num_workers > 0** What should i do next?
A tensor : `tensor([1., 1., 1., ..., 1., 1., 1.])`
@rusty1s After i add some parameter to this add_self_loops : ```python add_self_loops(edge_index = edge_index, num_nodes=num_nodes, edge_attr=edge_weight, fill_value=1) ``` The error become like this : `RuntimeError: expand(torch.LongTensor{[2, 30947]}, size=[30947]): the number...
``` edge_weight shape : torch.Size([30947]) row shape: torch.Size([2, 30947]) ```
edge_index : (tensor([[ 0, 0, 0, ..., 1180, 1181, 1182], [ 200, 201, 202, ..., 1180, 1181, 1182]]), None) edge_index is a tuple after inputed to add_self_loops