size mismatch for pos_embed: copying a param with shape torch.Size([1, 14, 14, 768]) from checkpoint, the shape in current model is torch.Size([1, 14, 14, 512]).
zhang-pan opened this issue · comments
When I use the pre-trained model d5-448, the following error appears:
Traceback (most recent call last):
File "F:/volo-main/main1_all_complete.py", line 415, in
main()
File "F:/volo-main/main1_all_complete.py", line 72, in main
load_pretrained_weights(model, './path/to/pretrained/weights/d5_448_87.0.pth.tar', use_ema=False, strict=False,num_classes=1000)
File "F:\volo-main\utils\utils.py", line 142, in load_pretrained_weights
model.load_state_dict(state_dict, strict=strict)
File "D:\Python36\lib\site-packages\torch\nn\modules\module.py", line 1224, in load_state_dict
self.class.name, "\n\t".join(error_msgs)))
RuntimeError: Error(s) in loading state_dict for VOLO:
size mismatch for pos_embed: copying a param with shape torch.Size([1, 14, 14, 768]) from checkpoint, the shape in current model is torch.Size([1, 14, 14, 512]).
size mismatch for cls_token: copying a param with shape torch.Size([1, 1, 768]) from checkpoint, the shape in current model is torch.Size([1, 1, 512]).
size mismatch for patch_embed.conv.0.weight: copying a param with shape torch.Size([128, 3, 7, 7]) from checkpoint, the shape in current model is torch.Size([64, 3, 7, 7]).
size mismatch for patch_embed.conv.1.weight: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.conv.1.bias: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.conv.1.running_mean: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.conv.1.running_var: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.conv.3.weight: copying a param with shape torch.Size([128, 128, 3, 3]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for patch_embed.conv.4.weight: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.conv.4.bias: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.conv.4.running_mean: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.conv.4.running_var: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.conv.6.weight: copying a param with shape torch.Size([128, 128, 3, 3]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for patch_embed.conv.7.weight: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.conv.7.bias: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.conv.7.running_mean: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.conv.7.running_var: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for patch_embed.proj.weight: copying a param with shape torch.Size([384, 128, 4, 4]) from checkpoint, the shape in current model is torch.Size([256, 64, 4, 4]).
size mismatch for patch_embed.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.0.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.0.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.0.attn.v.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.0.attn.attn.weight: copying a param with shape torch.Size([972, 384]) from checkpoint, the shape in current model is torch.Size([648, 256]).
size mismatch for network.0.0.attn.attn.bias: copying a param with shape torch.Size([972]) from checkpoint, the shape in current model is torch.Size([648]).
size mismatch for network.0.0.attn.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.0.attn.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.0.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.0.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.0.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
size mismatch for network.0.0.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for network.0.0.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([256, 768]).
size mismatch for network.0.0.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.1.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.1.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.1.attn.v.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.1.attn.attn.weight: copying a param with shape torch.Size([972, 384]) from checkpoint, the shape in current model is torch.Size([648, 256]).
size mismatch for network.0.1.attn.attn.bias: copying a param with shape torch.Size([972]) from checkpoint, the shape in current model is torch.Size([648]).
size mismatch for network.0.1.attn.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.1.attn.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.1.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.1.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.1.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
size mismatch for network.0.1.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for network.0.1.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([256, 768]).
size mismatch for network.0.1.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.2.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.2.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.2.attn.v.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.2.attn.attn.weight: copying a param with shape torch.Size([972, 384]) from checkpoint, the shape in current model is torch.Size([648, 256]).
size mismatch for network.0.2.attn.attn.bias: copying a param with shape torch.Size([972]) from checkpoint, the shape in current model is torch.Size([648]).
size mismatch for network.0.2.attn.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.2.attn.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.2.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.2.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.2.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
size mismatch for network.0.2.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for network.0.2.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([256, 768]).
size mismatch for network.0.2.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.3.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.3.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.3.attn.v.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.3.attn.attn.weight: copying a param with shape torch.Size([972, 384]) from checkpoint, the shape in current model is torch.Size([648, 256]).
size mismatch for network.0.3.attn.attn.bias: copying a param with shape torch.Size([972]) from checkpoint, the shape in current model is torch.Size([648]).
size mismatch for network.0.3.attn.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.3.attn.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.3.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.3.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.3.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
size mismatch for network.0.3.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for network.0.3.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([256, 768]).
size mismatch for network.0.3.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.4.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.4.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.4.attn.v.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.4.attn.attn.weight: copying a param with shape torch.Size([972, 384]) from checkpoint, the shape in current model is torch.Size([648, 256]).
size mismatch for network.0.4.attn.attn.bias: copying a param with shape torch.Size([972]) from checkpoint, the shape in current model is torch.Size([648]).
size mismatch for network.0.4.attn.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.4.attn.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.4.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.4.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.4.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
size mismatch for network.0.4.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for network.0.4.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([256, 768]).
size mismatch for network.0.4.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.5.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.5.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.5.attn.v.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.5.attn.attn.weight: copying a param with shape torch.Size([972, 384]) from checkpoint, the shape in current model is torch.Size([648, 256]).
size mismatch for network.0.5.attn.attn.bias: copying a param with shape torch.Size([972]) from checkpoint, the shape in current model is torch.Size([648]).
size mismatch for network.0.5.attn.proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
size mismatch for network.0.5.attn.proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.5.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.5.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.0.5.mlp.fc1.weight: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
size mismatch for network.0.5.mlp.fc1.bias: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([768]).
size mismatch for network.0.5.mlp.fc2.weight: copying a param with shape torch.Size([384, 1536]) from checkpoint, the shape in current model is torch.Size([256, 768]).
size mismatch for network.0.5.mlp.fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
size mismatch for network.1.proj.weight: copying a param with shape torch.Size([768, 384, 2, 2]) from checkpoint, the shape in current model is torch.Size([512, 256, 2, 2]).
size mismatch for network.1.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.0.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.0.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.0.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.2.0.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.2.0.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.0.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.0.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.0.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.2.0.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.2.0.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.2.0.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.1.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.1.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.1.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.2.1.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.2.1.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.1.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.1.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.1.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.2.1.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.2.1.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.2.1.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.2.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.2.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.2.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.2.2.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.2.2.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.2.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.2.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.2.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.2.2.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.2.2.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.2.2.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.3.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.3.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.3.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.2.3.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.2.3.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.3.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.3.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.2.3.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.2.3.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.2.3.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.2.3.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.0.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.0.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.0.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.0.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.3.0.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.0.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.0.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.0.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.0.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.3.0.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.3.0.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.1.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.1.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.1.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.1.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.3.1.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.1.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.1.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.1.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.1.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.3.1.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.3.1.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.2.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.2.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.2.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.2.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.3.2.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.2.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.2.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.2.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.2.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.3.2.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.3.2.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.3.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.3.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.3.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.3.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.3.3.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.3.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.3.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.3.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.3.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.3.3.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.3.3.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.4.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.4.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.4.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.4.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.3.4.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.4.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.4.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.4.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.4.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.3.4.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.3.4.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.5.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.5.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.5.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.5.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.3.5.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.5.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.5.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.5.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.5.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.3.5.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.3.5.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.6.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.6.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.6.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.6.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.3.6.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.6.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.6.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.6.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.6.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.3.6.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.3.6.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.7.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.7.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.7.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.7.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.3.7.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.7.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.7.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.7.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.7.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.3.7.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.3.7.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.8.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.8.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.8.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.8.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.3.8.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.8.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.8.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.8.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.8.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.3.8.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.3.8.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.9.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.9.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.9.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.9.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.3.9.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.9.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.9.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.3.9.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.3.9.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.3.9.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.3.9.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.0.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.0.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.0.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.4.0.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.4.0.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.0.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.0.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.0.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.4.0.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.4.0.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.4.0.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.1.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.1.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.1.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.4.1.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.4.1.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.1.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.1.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.1.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.4.1.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.4.1.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.4.1.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.2.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.2.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.2.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.4.2.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.4.2.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.2.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.2.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.2.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.4.2.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.4.2.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.4.2.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.3.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.3.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.3.attn.qkv.weight: copying a param with shape torch.Size([2304, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.4.3.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for network.4.3.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.3.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.3.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for network.4.3.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for network.4.3.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for network.4.3.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for network.4.3.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.0.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.0.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.0.attn.kv.weight: copying a param with shape torch.Size([1536, 768]) from checkpoint, the shape in current model is torch.Size([1024, 512]).
size mismatch for post_network.0.attn.q.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for post_network.0.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for post_network.0.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.0.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.0.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.0.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for post_network.0.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for post_network.0.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for post_network.0.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.1.norm1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.1.norm1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.1.attn.kv.weight: copying a param with shape torch.Size([1536, 768]) from checkpoint, the shape in current model is torch.Size([1024, 512]).
size mismatch for post_network.1.attn.q.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for post_network.1.attn.proj.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([512, 512]).
size mismatch for post_network.1.attn.proj.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.1.norm2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.1.norm2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for post_network.1.mlp.fc1.weight: copying a param with shape torch.Size([3072, 768]) from checkpoint, the shape in current model is torch.Size([1536, 512]).
size mismatch for post_network.1.mlp.fc1.bias: copying a param with shape torch.Size([3072]) from checkpoint, the shape in current model is torch.Size([1536]).
size mismatch for post_network.1.mlp.fc2.weight: copying a param with shape torch.Size([768, 3072]) from checkpoint, the shape in current model is torch.Size([512, 1536]).
size mismatch for post_network.1.mlp.fc2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for aux_head.weight: copying a param with shape torch.Size([1000, 768]) from checkpoint, the shape in current model is torch.Size([1000, 512]).
size mismatch for norm.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for norm.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for head.weight: copying a param with shape torch.Size([1000, 768]) from checkpoint, the shape in current model is torch.Size([1000, 512]).
did you solve it please ?