Error: bool value of Tensor with more than one value is ambiguous

[sarvesh710]

File "/home/sarvesh23/pytorch_RVAE/utils/functional.py", line 6, in f_and
return x and y
RuntimeError: bool value of Tensor with more than one value is ambiguous

I am running train_word_embeddings.py. Any hint what I am doing wrong ?

[davislf2]

Same problem here. Does anyone know why?

[SHIVITG]

preprocessed data was found and loaded
Traceback (most recent call last):
File "train_word_embeddings.py", line 50, in
out = neg_loss( input, target, args.num_sample).mean()
File "/home/ec2-user/anaconda3/envs/pytorch_p36/lib/python3.6/site-packages/torch/nn/modules/module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "/home/ec2-user/SageMaker/pytorch_RVAE/selfModules/neg.py", line 38, in forward
assert parameters_allocation_check(self),
File "/home/ec2-user/SageMaker/pytorch_RVAE/utils/functional.py", line 15, in parameters_allocation_check
return fold(f_and, parameters, True) or not fold(f_or, parameters, False)
File "/home/ec2-user/SageMaker/pytorch_RVAE/utils/functional.py", line 2, in fold
return a if (len(l) == 0) else fold(f, l[1:], f(a, l[0]))
File "/home/ec2-user/SageMaker/pytorch_RVAE/utils/functional.py", line 2, in fold
return a if (len(l) == 0) else fold(f, l[1:], f(a, l[0]))
File "/home/ec2-user/SageMaker/pytorch_RVAE/utils/functional.py", line 6, in f_and
return x and y
RuntimeError: bool value of Tensor with more than one value is ambiguous

[leehaoyuan]

According to the note, if you delete following lines in selfModules/neg.py, it will work just fine.
assert parameters_allocation_check(self),
"""
Invalid CUDA options. out_embed and in_embed parameters both should be stored in the same memory
got out_embed.is_cuda = {}, in_embed.is_cuda = {}
""".format(self.out_embed.weight.is_cuda, self.in_embed.weight.is_cuda)

[kay312]

According to the note, if you delete following lines in selfModules/neg.py, it will work just fine.
assert parameters_allocation_check(self),
"""
Invalid CUDA options. out_embed and in_embed parameters both should be stored in the same memory
got out_embed.is_cuda = {}, in_embed.is_cuda = {}
""".format(self.out_embed.weight.is_cuda, self.in_embed.weight.is_cuda)

yeah, the code is to do 'parameters_allocation_check', i deleted them and it worked, but i dont know whether it influence the output or not.

[gohjiayi]

According to the note, if you delete following lines in selfModules/neg.py, it will work just fine.
assert parameters_allocation_check(self),
"""
Invalid CUDA options. out_embed and in_embed parameters both should be stored in the same memory
got out_embed.is_cuda = {}, in_embed.is_cuda = {}
""".format(self.out_embed.weight.is_cuda, self.in_embed.weight.is_cuda)

Currently working on Python 3.6.9 and facing the same issue. After removing the parameters_allocation_check code (quoted above), I faced additional errors and this is how I solved them. (P.S. line number might differ)

ValueError: 'Object arrays cannot be loaded when allow_pickle=False'
In batch_loader.py, line 221. For np.load add in the argument allow_pickle=True, as instructed by StackOverflow post here.

[self.word_tensor, self.character_tensor] = [np.array([np.load(target, allow_pickle=True) for target in input_type])
                                            for input_type in tensor_files]

IndexError: too many indices for array: array is 0-dimensional, but 1 were indexed
In train_word_embeddings.py, line 56. For out.cpu().data.numpy()[0] remove the index [0].

if iteration % 500 == 0:
      out = out.cpu().data.numpy()
      print('iteration = {}, loss = {}'.format(iteration, out))

This allowed me to run the codes and build a custom word embedding successfully. Still studying the impact it has on the word embeddings so please use at your own discretion. Hope this helps the others!