Hi Guys,
I have a pre-trained resnet-50 model, which upon training gives this error: The size of tensor a (1000) must match the size of tensor b (10) at non-singleton dimension 1.
I am using MSELoss.
for epoch in range(n_epochs):
N = len(train_loader)
epoch_train_loss, epoch_test_loss = 0, 0
val_age_mse, ctr = 0, 0
_n = len(train_loader)
for ix, data in enumerate(train_loader):
# if ix == 100: break
**loss = train_batch(data, model, optimizer, criteria)** (This is where error shows up)
epoch_train_loss += loss.item()
This is where error crops up:
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
<ipython-input-103-01cf2dc1c2bf> in <module>()
30 for ix, data in enumerate(train_loader):
31 # if ix == 100: break
---> 32 loss = train_batch(data, model, optimizer, criteria)
33 epoch_train_loss += loss.item()
34
4 frames
/usr/local/lib/python3.7/dist-packages/torch/functional.py in broadcast_tensors(*tensors)
69 if any(type(t) is not Tensor for t in tensors) and has_torch_function(tensors):
70 return handle_torch_function(broadcast_tensors, tensors, *tensors)
---> 71 return _VF.broadcast_tensors(tensors) # type: ignore