Closed ShafinH closed 2 weeks ago
Can you try using the new download_weights script? Should reduce these types of errors.
Is there anyway to get the weights that were used when the infer.py file used to exist? I'm trying to run the version of the code on https://github.com/victorchall/genmoai-smol, but it seems like the code was refactored quite a bit recently.
The official repository supports running on a single GPU now, so I'd recommend checking out the latest commit from this repo :)
Closing this, feel free to leave a comment if it's still not working.
Downloaded weights from the hugging face link
RuntimeError: Error(s) in loading state_dict for AsymmDiTJoint: Unexpected key(s) in state_dict: "t5_y_embedder.to_kv.bias", "t5_y_embedder.to_kv.weight", "t5_y_embedder.to_out.bias", "t5_y_embedder.to_out.weight", "t5_y_embedder.to_q.bias", "t5_y_embedder.to_q.weight", "t5_yproj.bias", "t5_yproj.weight". size mismatch for final_layer.linear.weight: copying a param with shape torch.Size([48, 3072]) from checkpoint, the shape in current model is torch.Size([96, 3072]). size mismatch for final_layer.linear.bias: copying a param with shape torch.Size([48]) from checkpoint, the shape in current model is torch.Size([96]).