RuntimeError: The shape of the 2D attn_mask is torch.Size([77, 77]), but should be (1, 1).

[rajatanand86]

Gettin below error: Can someboday help? I am running python i2v_test.py

with torch.no_grad(), torch.cuda.amp.autocast(): Traceback (most recent call last): File "/home/ec2-user/SageMaker/DynamiCrafter/i2v_test.py", line 146, in video_path = i2v.get_image('prompts/art.png','man fishing in a boat at sunset') ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/SageMaker/DynamiCrafter/i2v_test.py", line 65, in get_image text_emb = model.get_learned_conditioning([prompt]) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/SageMaker/DynamiCrafter/lvdm/models/ddpm3d.py", line 601, in get_learned_conditioning c = self.cond_stage_model.encode(c) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/SageMaker/DynamiCrafter/lvdm/modules/encoders/condition.py", line 234, in encode return self(text) ^^^^^^^^^^ File "/home/ec2-user/anaconda3/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1736, in _wrapped_call_impl return self._call_impl(*args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/anaconda3/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1747, in _call_impl return forward_call(*args, *kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/SageMaker/DynamiCrafter/lvdm/modules/encoders/condition.py", line 211, in forward z = self.encode_with_transformer(tokens.to(self.device)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/SageMaker/DynamiCrafter/lvdm/modules/encoders/condition.py", line 218, in encode_with_transformer x = self.text_transformer_forward(x, attn_mask=self.model.attn_mask) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/SageMaker/DynamiCrafter/lvdm/modules/encoders/condition.py", line 230, in text_transformer_forward x = r(x, attn_mask=attn_mask) ^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/anaconda3/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1736, in _wrapped_call_impl return self._call_impl(args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/anaconda3/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1747, in _call_impl return forward_call(*args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/anaconda3/lib/python3.12/site-packages/open_clip/transformer.py", line 263, in forward x = q_x + self.ls_1(self.attention(q_x=self.ln_1(q_x), k_x=k_x, v_x=v_x, attn_mask=attn_mask)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/anaconda3/lib/python3.12/site-packages/open_clip/transformer.py", line 250, in attention return self.attn( ^^^^^^^^^^ File "/home/ec2-user/anaconda3/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1736, in _wrapped_call_impl return self._call_impl(*args, *kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/anaconda3/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1747, in _call_impl return forward_call(args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/anaconda3/lib/python3.12/site-packages/torch/nn/modules/activation.py", line 1368, in forward attn_output, attn_output_weights = F.multi_head_attention_forward( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/ec2-user/anaconda3/lib/python3.12/site-packages/torch/nn/functional.py", line 6131, in multi_head_attention_forward raise RuntimeError( RuntimeError: The shape of the 2D attn_mask is torch.Size([77, 77]), but should be (1, 1).

[ascv0228]

same problem

[Doubiiu]

Could you show your package versions, especially for open_clip and transformer? There should be some problems here.