Problem encountered in train.py

[MUrielleMU]

Excuse me, I met this problem running train.py.
Could you plz tell me if I did anything wrong?
I really appreciate your reply :)

File "../aspect_category_model/bert_capsnet.py", line 65, in forward
sentence_mask = segment_mask & (1 - bert_segment).byte()
RuntimeError: result type Byte can't be cast to the desired output type Bool

[jiangqn]

bert_segment should be a LongTensor. Maybe the type of input got wrong.

[bujiahao]

Excuse me, I met this problem running train.py.
Could you plz tell me if I did anything wrong?
I really appreciate your reply :)

File "../aspect_category_model/bert_capsnet.py", line 65, in forward
sentence_mask = segment_mask & (1 - bert_segment).byte()
RuntimeError: result type Byte can't be cast to the desired output type Bool

I also met the same problem when run train.py
sentence_mask = segment_mask & (1 - bert_segment).byte()
RuntimeError: Expected object of scalar type Bool but got scalar type Byte for argument #2 'other'

do u solve this problem?

[bujiahao]

oh
version donot match

[C1oudysTarT]

oh
version donot match

Could you tell me how to solve this problem? Thanks a lot.

[scofield7419]

sentence_mask = segment_mask & (1 - bert_segment).byte()
->
sentence_mask = segment_mask & (1 - bert_segment).bool()

[mhaber]

If you use torch=1.1.0 then it should work.