Expected object of scalar type Bool but got scalar type Byte for argument #2

im trying to use BatchHardTripletLoss you provided as Loss function in combination with SentenceLabelDataset (it works without any issue when i use the normal TripletLoss)

when i start the fit method, i get the error:

`RuntimeError: Expected object of scalar type Bool but got scalar type Byte for argument #2 'other' in call to _th_and`

it occurs at _get_anchor_positive_triplet_mask method, and the whole dump is like:

```
raceback (most recent call last): | 0/152 [00:00
cli.main()
File "c:\Users\tyker1\.vscode\extensions\ms-python.python-2020.3.69010\pythonFiles\lib\python\debugpy\wheels\debugpy/..\debugpy\server\cli.py", line 427, in main
run()
File "c:\Users\tyker1\.vscode\extensions\ms-python.python-2020.3.69010\pythonFiles\lib\python\debugpy\wheels\debugpy/..\debugpy\server\cli.py", line 264, in run_file
runpy.run_path(options.target, run_name="__main__")
File "C:\Users\tyker1\AppData\Local\Programs\Python\Python37\lib\runpy.py", line 263, in run_path
pkg_name=pkg_name, script_name=fname)
File "C:\Users\tyker1\AppData\Local\Programs\Python\Python37\lib\runpy.py", line 96, in _run_module_code
mod_name, mod_spec, pkg_name, script_name)
File "C:\Users\tyker1\AppData\Local\Programs\Python\Python37\lib\runpy.py", line 85, in _run_code
exec(code, run_globals)
File "d:\MeinProgramme\MeinPython\Masterarbeit\pythonCodes\siameseTest\sbert-test.py", line 56, in
fp16_opt_level='O2')
File "C:\Users\tyker1\AppData\Local\Programs\Python\Python37\lib\site-packages\sentence_transformers\SentenceTransformer.py", line 351, in fit
loss_value = loss_model(features, labels)
File "C:\Users\tyker1\AppData\Local\Programs\Python\Python37\lib\site-packages\torch\nn\modules\module.py", line 532, in __call__
result = self.forward(*input, **kwargs)
File "C:\Users\tyker1\AppData\Local\Programs\Python\Python37\lib\site-packages\apex\amp\_initialize.py", line 197, in new_fwd
**applier(kwargs, input_caster))
File "C:\Users\tyker1\AppData\Local\Programs\Python\Python37\lib\site-packages\sentence_transformers\losses\BatchHardTripletLoss.py", line 15, in forward
return BatchHardTripletLoss.batch_hard_triplet_loss(labels, reps[0], margin=self.triplet_margin)
File "C:\Users\tyker1\AppData\Local\Programs\Python\Python37\lib\site-packages\sentence_transformers\losses\BatchHardTripletLoss.py", line 39, in batch_hard_triplet_loss
mask_anchor_positive = BatchHardTripletLoss._get_anchor_positive_triplet_mask(labels).float()
File "C:\Users\tyker1\AppData\Local\Programs\Python\Python37\lib\site-packages\sentence_transformers\losses\BatchHardTripletLoss.py", line 188, in _get_anchor_positive_triplet_mask
return labels_equal & indices_not_equal
RuntimeError: Expected object of scalar type Bool but got scalar type Byte for argument #2 'other' in call to _th_and
```

Start in sentence_transformers/losses/BatchHardTripletLoss.py, especially _get_anchor_positive_triplet_mask and the batch_hard_triplet_loss call shown in the traceback. Reproduce the failure with BatchHardTripletLoss and SentenceLabelDataset during fit; done when this combination runs without the Bool/Byte runtime error while normal TripletLoss behavior remains intact.

Written by the indexing model from the issue text.