A problem that checkpoint doesn't match with the model #17
Description
I met the warning output like this. But the overall pipeline functions well and the demo's output is right.
Some weights of UNet2DConditionModel_main were not initialized from the model checkpoint at ./checkpoints/marigold-depth-v1-0 and are newly initialized because the shapes did not match:
- conv_in.weight: found shape torch.Size([320, 8, 3, 3]) in the checkpoint and torch.Size([320, 12, 3, 3]) in the model instantiated
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
Some weights of the model checkpoint at ./checkpoints/marigold-depth-v1-0 were not used when initializing UNet2DConditionModel: ['conv_out.bias', 'conv_norm_out.weight', 'conv_out.weight', 'conv_norm_out.bias']
- This IS expected if you are initializing UNet2DConditionModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
- This IS NOT expected if you are initializing UNet2DConditionModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
Some weights of UNet2DConditionModel were not initialized from the model checkpoint at ./checkpoints/marigold-depth-v1-0 and are newly initialized because the shapes did not match:
- conv_in.weight: found shape torch.Size([320, 8, 3, 3]) in the checkpoint and torch.Size([320, 4, 3, 3]) in the model instantiated
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference
