BertGCN

BertGCN copied to clipboard

build_graph中shuffle部分好像有数据泄露

6

ydwasd

TypeError: GraphConv.forward() got an unexpected keyword argument 'edge_weights'

3

ERROR:ignite.engine.engine.Engine:Current run is terminating due to exception: GraphConv.forward() got an unexpected keyword argument 'edge_weights' ERROR:ignite.engine.engine.Engine:Engine run is terminating due to exception: GraphConv.forward() got an unexpected keyword argument 'edge_weights' --------------------------------------------------------------------------- TypeError...

NitishOritro

ModuleNotFoundError: No module named 'scipy.sparse.linalg.eigen.arpack'; 'scipy.sparse.linalg.eigen' is not a package

2

Hi, I just run your code `build_graph.py` but it appears the issues ModuleNotFound although I've created Anaconda environment and installed Scipy package. Can you please tell me why this error...

MarkusKhoa

Without proper garbage collection, building graphs on a large dataset required a lot of memory. With `gc` I was able to bring it down by a lot. I have also...

FrozenWolf-Cyber

CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasSgemm( handle, opa, opb, m, n, k, &alpha, a, lda, b, ldb, &beta, c, ldc)`

luckyonetwo

Hi @ZeroRin, I want to train the model on a custom dataset, Could you please help me with the format of dataset for the 2nd and 3rd scripts. For the...

Sagar1094

@ZeroRin Just averaging the loss of in one batche may be confusing for the following reasons: 1) Since the number of training samples in one batch are not always the...

qhd1996

Computation Requirements

1

We have 2.8M data. Would it be feasible to train your architecture on this large data? If yes, how much memory requirements and time would be needed for it approx?

UditArora2000

cpu

1

can I only use cpu to run the code?

chenkaixin66

CUDA out of memory.

6

Is it very demanding on the GPU?I use RTX3090 to run the code, it will show a memory overflow

dywe666