simonucl / ldgn-mltc Goto Github PK

When i run the train.py for 10 epoch, I successfully run the first epoch but went wrong when running the second epoch.
Here, I post the wrong text:

return torch._C._nn.binary_cross_entropy(input, target, weight, reduction_enum)
RuntimeError: all elements of input should be between 0 and 1

How can i solve this problem?
Thanks very much.

Hi, could I know the number of epoch you run? since I can't find it in paper and in code it was set to 10, so if it only need 10 epochs enough?
It runs successfully for the first few epochs, but after running a few more epochs, some errors may appear randomly. Sometimes in the tenth epoch and sometimes in the twentieth, it suddenly went wrong and gave me a RuntimeError:
RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling cublasSgemmStridedBatched( handle, opa, opb, m, n, k, &alpha, a, lda, stridea, b, ldb, strideb, &beta, c, ldc, stridec, num_batches)
or
RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling cublasSgemm( handle, opa, opb, m, n, k, &alpha, a, lda, b, ldb, &beta, c, ldc)
Could you give me some advice? Thanks.