人像遮罩 Ultra V2

[ok1248]

Error(s) in loading state_dict for VitMatteForImageMatting:
size mismatch for backbone.embeddings.position_embeddings: copying a param with shape torch.Size([1, 197, 384]) from checkpoint, the shape in current model is torch.Size([1, 197, 768]).
size mismatch for backbone.embeddings.projection.weight: copying a param with shape torch.Size([384, 4, 16, 16]) from checkpoint, the shape in current model is torch.Size([768, 3, 16, 16]).
size mismatch for backbone.embeddings.projection.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.0.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.0.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.0.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.0.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.0.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.0.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.0.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.0.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.0.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.0.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.0.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.0.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.1.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.1.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.1.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.1.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.1.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.1.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.1.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.1.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.1.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.1.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.1.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.1.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.2.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.2.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.2.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.2.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.2.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.2.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.2.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.2.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.2.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.2.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.2.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.2.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.3.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.3.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.3.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.3.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.3.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.3.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.3.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.3.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.3.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.3.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.3.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.3.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.4.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.4.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.4.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.4.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.4.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.4.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.4.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.4.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.4.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.4.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.4.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.4.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.5.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.5.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.5.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.5.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.5.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.5.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.5.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.5.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.5.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.5.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.5.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.5.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.6.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.6.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.6.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.6.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.6.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.6.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.6.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.6.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.6.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.6.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.6.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.6.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.7.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.7.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.7.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.7.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.7.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.7.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.7.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.7.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.7.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.7.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.7.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.7.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.8.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.8.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.8.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.8.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.8.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.8.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.8.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.8.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.8.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.8.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.8.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.8.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.9.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.9.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.9.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.9.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.9.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.9.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.9.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.9.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.9.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.9.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.9.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.9.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.10.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.10.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.10.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.10.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.10.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.10.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.10.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.10.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.10.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.10.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.10.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.10.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.11.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.11.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.11.attention.qkv.weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([2304, 768]).
size mismatch for backbone.encoder.layer.11.attention.qkv.bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([2304]).
size mismatch for backbone.encoder.layer.11.attention.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([768, 768]).
size mismatch for backbone.encoder.layer.11.attention.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.11.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.11.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for backbone.encoder.layer.11.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([3072, 768]).
size mismatch for backbone.encoder.layer.11.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([3072]).
size mismatch for backbone.encoder.layer.11.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([768, 3072]).
size mismatch for backbone.encoder.layer.11.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for decoder.convstream.convs.0.conv.weight: copying a param with shape torch.Size([48, 4, 3, 3]) from checkpoint, the shape in current model is torch.Size([48, 3, 3, 3]).
size mismatch for decoder.fusion_blocks.3.conv.conv.weight: copying a param with shape torch.Size([32, 68, 3, 3]) from checkpoint, the shape in current model is torch.Size([32, 67, 3, 3]).
You may consider adding ignore_mismatched_sizes=True in the model from_pretrained method.