RuntimeError: The size of tensor a (3) must match the size of tensor b (9) at non-singleton dimension 0

[studiumcirclus]

All of my images are 256x256 pixels (taken from images sample folder, just some of the 256x256 ones)

C:\git\image-captioning>python inference.py -i C:\git\image-captioning\inputs --batch 3 --gpu 0
Device: cpu
Images found: 8
Split size: 2
Checkpoint loading...
load checkpoint from ./checkpoints/model_large_caption.pth

Model to cpu
Inference started
0batch [00:02, ?batch/s]
Traceback (most recent call last):
File "C:\git\image-captioning\inference.py", line 88, in
caption = model.generate(
File "C:\git\image-captioning\models\blip.py", line 201, in generate
outputs = self.text_decoder.generate(
File "C:\Python\Python310\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "C:\Python\Python310\lib\site-packages\transformers\generation\utils.py", line 1752, in generate
return self.beam_search(
File "C:\Python\Python310\lib\site-packages\transformers\generation\utils.py", line 3091, in beam_search
outputs = self(
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "C:\git\image-captioning\models\med.py", line 886, in forward
outputs = self.bert(
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "C:\git\image-captioning\models\med.py", line 781, in forward
encoder_outputs = self.encoder(
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "C:\git\image-captioning\models\med.py", line 445, in forward
layer_outputs = layer_module(
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "C:\git\image-captioning\models\med.py", line 361, in forward
cross_attention_outputs = self.crossattention(
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "C:\git\image-captioning\models\med.py", line 277, in forward
self_outputs = self.self(
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "C:\Python\Python310\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "C:\git\image-captioning\models\med.py", line 178, in forward
attention_scores = torch.matmul(query_layer, key_layer.transpose(-1, -2))
RuntimeError: The size of tensor a (3) must match the size of tensor b (9) at non-singleton dimension 0

[cobanov]

same with #5