Meeting issues during Evaluation

[lulinliu]

Hi, thank u for your great work.

I am meeting some errors when trying to evaluate the re10k. I am using this command: python3 -m src.main +experiment=re10k mode=test dataset/view_sampler=evaluation dataset.view_sampler.index_path=assets/evaluation_index_re10k.json checkpointing.load=checkpoints/re10k.ckpt And meeting this error: Loading model from: /opt/miniconda3/envs/mvsplat/lib/python3.10/site-packages/lpips/weights/v0.1/vgg.pth You are using a CUDA device ('NVIDIA A100 80GB PCIe') that has Tensor Cores. To properly utilize them, you should set torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision Restoring states from the checkpoint path at checkpoints/re10k.ckpt [2024-12-13 22:35:55,916][py.warnings][WARNING] - /opt/miniconda3/envs/mvsplat/lib/python3.10/site-packages/lightning/pytorch/utilities/migration/utils.py:55: The loaded checkpoint was produced with Lightning v2.2.0.post0, which is newer than your current Lightning version: v2.1.4

Error executing job with overrides: ['+experiment=re10k', 'mode=test', 'dataset/view_sampler=evaluation', 'dataset.view_sampler.index_path=assets/evaluation_index_re10k.json', 'checkpointing.load=checkpoints/re10k.ckpt'] Traceback (most recent call last): File "/datadrive/pixelsplat/src/main.py", line 130, in train trainer.test( File "/opt/miniconda3/envs/mvsplat/lib/python3.10/site-packages/lightning/pytorch/trainer/trainer.py", line 754, in test `

Also coming with this:

RuntimeError: Error(s) in loading state_dict for ModelWrapper: Missing key(s) in state_dict: "encoder.backbone.dino.cls_token", "encoder.backbone.dino.pos_embed", "encoder.backbone.dino.patch_embed.proj.weight", "encoder.backbone.dino.patch_embed.proj.bias", "encoder.backbone.dino.blocks.0.norm1.weight", "encoder.backbone.dino.blocks.0.norm1.bias", "encoder.backbone.dino.blocks.0.attn.qkv.weight", "encoder.backbone.dino.blocks.0.attn.qkv.bias", "encoder.backbone.dino.blocks.0.attn.proj.weight", "encoder.backbone.dino.blocks.0.attn.proj.bias", "encoder.backbone.dino.blocks.0.norm2.weight", "encoder.backbone.dino.blocks.0.norm2.bias", "encoder.backbone.dino.blocks.0.mlp.fc1.weight", "encoder.backbone.dino.blocks.0.mlp.fc1.bias", "encoder.backbone.dino.blocks.0.mlp.fc2.weight", "encoder.backbone.dino.blocks.0.mlp.fc2.bias", "encoder.backbone.dino.blocks.1.norm1.weight", "encoder.backbone.dino.blocks.1.norm1.bias", "encoder.backbone.dino.blocks.1.attn.qkv.weight", and lots of this information.

Do you know where I am wrong? Thank your in advance!