Lepard: Learning Partial point cloud matching in Rigid and Deformable scenes

in main.py line 99 ,'cofig.train_loader, neighborhood_limits = get_dataloader(train_set,config,shuffle=True)' this line have one problem. the 'train_set' need one file '3dmatch/3dmatch/train/rgbd-scenes-v2-scene_01/cloud_bin_1.pth' , but i can't find the ''cloud_bin_1.pth''. i have already download 3dmatch.zip ,and unzip this file. but in the unzip file,i don't find the ''cloud_bin_1.pth''. i hope you can help me. thank you very much.

Hello! Here are some questions for the code part on dataset 3DMatch.

1. The dimensionality of the output from the KPConv backbone is set to 528, theoretically, any number that could be divided by 6 and 4 with no remainder is feasible here? Because in my understanding, divided by 6 is for the rotary positional embedding, and divided by 4 is for the multi-head attention, so why not choose 516 as the output dimensionality? Is there any reason behind this choice?
2. The code part of rotary positional embedding you implemented is impressive, I noticed that you first voxelize the raw 3d coordinates (but not flooring the output coordinates) to scale the 3d coordinates, then I believe the vol_bnds = [-3.6, -2.4, 1.14] is the minimal coordinates among all 3DMatch train&val&test set? also, does voxelizing(or called scaling) the coordinates before positional embedding could get better results than just utilizing the raw 3d coordinates for positional embedding? Could you give some hints about it?

Thank you very much for your help.

(Thanks for your excellent work bro ! this paper is really nice and amazing~ When i try to train the lepard myself ( I only change the "batchsize=1" and "num_workers=0"), i found the memory is keep raising(i have 126g mem, lepard almost cost 100g and still raising ), finally it was killed by linux

Have you encountered this kind of problem before? I did not change your code, i guess this problem come from dataloader? can you give me some suggestions ?

Hi Yang, how are EPE3D, Acc5 and Acc10 computed on 4DMatch? Are they computed on ALL points in the source point cloud or only the points from "metric_index" in the dataset?

Hi Yang,

I tried to run the pretrained weights on 4DMatch/4DLoMatch, but the results are lower than those reported in the paper. Do I need more modifications to get the results in the paper?

Hi,

Thank you for sharing your excellent work! I have read the paper recently, and I have some questions about the position encoding. In your paper, you add position encoding to three places: self-attention, cross-attention, confidence matrix. And as you mention in the paper, position encoding in cross-attention and matching may mislead the registration at the beginning. So, I wonder if you have compared the performance at the situation you remove the position encoding at the cross-attention part and the confidence matrix part at the first stage but retain them at the second stage. Plus, I think the figures in your paper are very beautiful. So could you please tell me how you draw the point cloud and correspondences of figure 7 in your paper and the point cloud video in the readme? I will be very appreciative if you can give me some hints about the above question. Thank you very much!

thanks for sharing such inspiring work, i was trying to retest inference performance based on 3dmatch dataset and followed the instruction using this command to start testing: python main.py configs/test/3dmatch.yaml however inference procedure failed since the error below:

File "/home/hadley/Development/lepard/lib/trainer.py", line 80, in _load_pretrain
    if os.path.isfile(resume):
  File "/home/hadley/anaconda3/envs/torch_12/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1604, in load_state_dict
    raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for Pipeline:
	Missing key(s) in state_dict: "backbone.encoder_blocks.0.KPConv.weights", "backbone.encoder_blocks.0.KPConv.kernel_points", "backbone.encoder_blocks.1.unary1.mlp.weight", "backbone.encoder_blocks.1.KPConv.weights", "backbone.encoder_blocks.1.KPConv.kernel_points", "backbone.encoder_blocks.1.unary2.mlp.weight", "backbone.encoder_blocks.1.unary_shortcut.mlp.weight", "backbone.encoder_blocks.2.unary1.mlp.weight", "backbone.encoder_blocks.2.KPConv.weights", "backbone.encoder_blocks.2.KPConv.kernel_points", "backbone.encoder_blocks.2.unary2.mlp.weight", "backbone.encoder_blocks.3.unary1.mlp.weight", "backbone.encoder_blocks.3.KPConv.weights", "backbone.encoder_blocks.3.KPConv.kernel_points", "backbone.encoder_blocks.3.unary2.mlp.weight", "backbone.encoder_blocks.3.unary_shortcut.mlp.weight", "backbone.encoder_blocks.4.unary1.mlp.weight", "backbone.encoder_blocks.4.KPConv.weights", "backbone.encoder_blocks.4.KPConv.kernel_points", "backbone.encoder_blocks.4.unary2.mlp.weight", "backbone.encoder_blocks.5.unary1.mlp.weight", "backbone.encoder_blocks.5.KPConv.weights", "backbone.encoder_blocks.5.KPConv.kernel_points", "backbone.encoder_blocks.5.unary2.mlp.weight", "backbone.encoder_blocks.6.unary1.mlp.weight", "backbone.encoder_blocks.6.KPConv.weights", "backbone.encoder_blocks.6.KPConv.kernel_points", "backbone.encoder_blocks.6.unary2.mlp.weight", "backbone.encoder_blocks.6.unary_shortcut.mlp.weight", "backbone.encoder_blocks.7.unary1.mlp.weight", "backbone.encoder_blocks.7.KPConv.weights", "backbone.encoder_blocks.7.KPConv.kernel_points", "backbone.encoder_blocks.7.unary2.mlp.weight", "backbone.encoder_blocks.8.unary1.mlp.weight", "backbone.encoder_blocks.8.KPConv.weights", "backbone.encoder_blocks.8.KPConv.kernel_points", "backbone.encoder_blocks.8.unary2.mlp.weight", "backbone.encoder_blocks.9.unary1.mlp.weight", "backbone.encoder_blocks.9.KPConv.weights", "backbone.encoder_blocks.9.KPConv.kernel_points", "backbone.encoder_blocks.9.unary2.mlp.weight", "backbone.encoder_blocks.9.unary_shortcut.mlp.weight", "backbone.encoder_blocks.10.unary1.mlp.weight", "backbone.encoder_blocks.10.KPConv.weights", "backbone.encoder_blocks.10.KPConv.kernel_points", "backbone.encoder_blocks.10.unary2.mlp.weight", "backbone.coarse_out.weight", "backbone.coarse_out.bias", "backbone.coarse_in.weight", "backbone.coarse_in.bias", "backbone.decoder_blocks.1.mlp.weight", "backbone.decoder_blocks.3.mlp.weight", "backbone.decoder_blocks.5.mlp.weight", "backbone.fine_out.weight", "backbone.fine_out.bias", "coarse_transformer.layers.0.q_proj.weight", "coarse_transformer.layers.0.k_proj.weight", "coarse_transformer.layers.0.v_proj.weight", "coarse_transformer.layers.0.merge.weight", "coarse_transformer.layers.0.mlp.0.weight", "coarse_transformer.layers.0.mlp.2.weight", "coarse_transformer.layers.0.norm1.weight", "coarse_transformer.layers.0.norm1.bias", "coarse_transformer.layers.0.norm2.weight", "coarse_transformer.layers.0.norm2.bias", "coarse_transformer.layers.1.q_proj.weight", "coarse_transformer.layers.1.k_proj.weight", "coarse_transformer.layers.1.v_proj.weight", "coarse_transformer.layers.1.merge.weight", "coarse_transformer.layers.1.mlp.0.weight", "coarse_transformer.layers.1.mlp.2.weight", "coarse_transformer.layers.1.norm1.weight", "coarse_transformer.layers.1.norm1.bias", "coarse_transformer.layers.1.norm2.weight", "coarse_transformer.layers.1.norm2.bias", "coarse_transformer.layers.2.0.src_proj.weight", "coarse_transformer.layers.2.0.tgt_proj.weight", "coarse_transformer.layers.2.0.instNormLayer.weight", "coarse_transformer.layers.2.0.instNormLayer.bias", "coarse_transformer.layers.2.0.edgeNormLayer.weight", "coarse_transformer.layers.2.0.edgeNormLayer.bias", "coarse_transformer.layers.3.q_proj.weight", "coarse_transformer.layers.3.k_proj.weight", "coarse_transformer.layers.3.v_proj.weight", "coarse_transformer.layers.3.merge.weight", "coarse_transformer.layers.3.mlp.0.weight", "coarse_transformer.layers.3.mlp.2.weight", "coarse_transformer.layers.3.norm1.weight", "coarse_transformer.layers.3.norm1.bias", "coarse_transformer.layers.3.norm2.weight", "coarse_transformer.layers.3.norm2.bias", "coarse_transformer.layers.4.q_proj.weight", "coarse_transformer.layers.4.k_proj.weight", "coarse_transformer.layers.4.v_proj.weight", "coarse_transformer.layers.4.merge.weight", "coarse_transformer.layers.4.mlp.0.weight", "coarse_transformer.layers.4.mlp.2.weight", "coarse_transformer.layers.4.norm1.weight", "coarse_transformer.layers.4.norm1.bias", "coarse_transformer.layers.4.norm2.weight", "coarse_transformer.layers.4.norm2.bias", "coarse_matching.src_proj.weight", "coarse_matching.tgt_proj.weight", "coarse_matching.instNormLayer.weight", "coarse_matching.instNormLayer.bias", "coarse_matching.edgeNormLayer.weight", "coarse_matching.edgeNormLayer.bias". 
	Unexpected key(s) in state_dict: "kpf_encoder.encoder_blocks.0.KPConv.weights", "kpf_encoder.encoder_blocks.0.KPConv.kernel_points", "kpf_encoder.encoder_blocks.1.unary1.mlp.weight", "kpf_encoder.encoder_blocks.1.KPConv.weights", "kpf_encoder.encoder_blocks.1.KPConv.kernel_points", "kpf_encoder.encoder_blocks.1.unary2.mlp.weight", "kpf_encoder.encoder_blocks.1.unary_shortcut.mlp.weight", "kpf_encoder.encoder_blocks.2.unary1.mlp.weight", "kpf_encoder.encoder_blocks.2.KPConv.weights", "kpf_encoder.encoder_blocks.2.KPConv.kernel_points", "kpf_encoder.encoder_blocks.2.unary2.mlp.weight", "kpf_encoder.encoder_blocks.3.unary1.mlp.weight", "kpf_encoder.encoder_blocks.3.KPConv.weights", "kpf_encoder.encoder_blocks.3.KPConv.kernel_points", "kpf_encoder.encoder_blocks.3.unary2.mlp.weight", "kpf_encoder.encoder_blocks.3.unary_shortcut.mlp.weight", "kpf_encoder.encoder_blocks.4.unary1.mlp.weight", "kpf_encoder.encoder_blocks.4.KPConv.weights", "kpf_encoder.encoder_blocks.4.KPConv.kernel_points", "kpf_encoder.encoder_blocks.4.unary2.mlp.weight", "kpf_encoder.encoder_blocks.5.unary1.mlp.weight", "kpf_encoder.encoder_blocks.5.KPConv.weights", "kpf_encoder.encoder_blocks.5.KPConv.kernel_points", "kpf_encoder.encoder_blocks.5.unary2.mlp.weight", "kpf_encoder.encoder_blocks.6.unary1.mlp.weight", "kpf_encoder.encoder_blocks.6.KPConv.weights", "kpf_encoder.encoder_blocks.6.KPConv.kernel_points", "kpf_encoder.encoder_blocks.6.unary2.mlp.weight", "kpf_encoder.encoder_blocks.6.unary_shortcut.mlp.weight", "kpf_encoder.encoder_blocks.7.unary1.mlp.weight", "kpf_encoder.encoder_blocks.7.KPConv.weights", "kpf_encoder.encoder_blocks.7.KPConv.kernel_points", "kpf_encoder.encoder_blocks.7.unary2.mlp.weight", "kpf_encoder.encoder_blocks.8.unary1.mlp.weight", "kpf_encoder.encoder_blocks.8.KPConv.weights", "kpf_encoder.encoder_blocks.8.KPConv.kernel_points", "kpf_encoder.encoder_blocks.8.unary2.mlp.weight", "kpf_encoder.encoder_blocks.9.unary1.mlp.weight", "kpf_encoder.encoder_blocks.9.KPConv.weights", "kpf_encoder.encoder_blocks.9.KPConv.kernel_points", "kpf_encoder.encoder_blocks.9.unary2.mlp.weight", "kpf_encoder.encoder_blocks.9.unary_shortcut.mlp.weight", "kpf_encoder.encoder_blocks.10.unary1.mlp.weight", "kpf_encoder.encoder_blocks.10.KPConv.weights", "kpf_encoder.encoder_blocks.10.KPConv.kernel_points", "kpf_encoder.encoder_blocks.10.unary2.mlp.weight", "feat_proj.weight", "feat_proj.bias", "transformer_encoder.layers.0.self_attn.in_proj_weight", "transformer_encoder.layers.0.self_attn.in_proj_bias", "transformer_encoder.layers.0.self_attn.out_proj.weight", "transformer_encoder.layers.0.self_attn.out_proj.bias", "transformer_encoder.layers.0.multihead_attn.in_proj_weight", "transformer_encoder.layers.0.multihead_attn.in_proj_bias", "transformer_encoder.layers.0.multihead_attn.out_proj.weight", "transformer_encoder.layers.0.multihead_attn.out_proj.bias", "transformer_encoder.layers.0.linear1.weight", "transformer_encoder.layers.0.linear1.bias", "transformer_encoder.layers.0.linear2.weight", "transformer_encoder.layers.0.linear2.bias", "transformer_encoder.layers.0.norm1.weight", "transformer_encoder.layers.0.norm1.bias", "transformer_encoder.layers.0.norm2.weight", "transformer_encoder.layers.0.norm2.bias", "transformer_encoder.layers.0.norm3.weight", "transformer_encoder.layers.0.norm3.bias", "transformer_encoder.layers.1.self_attn.in_proj_weight", "transformer_encoder.layers.1.self_attn.in_proj_bias", "transformer_encoder.layers.1.self_attn.out_proj.weight", "transformer_encoder.layers.1.self_attn.out_proj.bias", "transformer_encoder.layers.1.multihead_attn.in_proj_weight", "transformer_encoder.layers.1.multihead_attn.in_proj_bias", "transformer_encoder.layers.1.multihead_attn.out_proj.weight", "transformer_encoder.layers.1.multihead_attn.out_proj.bias", "transformer_encoder.layers.1.linear1.weight", "transformer_encoder.layers.1.linear1.bias", "transformer_encoder.layers.1.linear2.weight", "transformer_encoder.layers.1.linear2.bias", "transformer_encoder.layers.1.norm1.weight", "transformer_encoder.layers.1.norm1.bias", "transformer_encoder.layers.1.norm2.weight", "transformer_encoder.layers.1.norm2.bias", "transformer_encoder.layers.1.norm3.weight", "transformer_encoder.layers.1.norm3.bias", "transformer_encoder.layers.2.self_attn.in_proj_weight", "transformer_encoder.layers.2.self_attn.in_proj_bias", "transformer_encoder.layers.2.self_attn.out_proj.weight", "transformer_encoder.layers.2.self_attn.out_proj.bias", "transformer_encoder.layers.2.multihead_attn.in_proj_weight", "transformer_encoder.layers.2.multihead_attn.in_proj_bias", "transformer_encoder.layers.2.multihead_attn.out_proj.weight", "transformer_encoder.layers.2.multihead_attn.out_proj.bias", "transformer_encoder.layers.2.linear1.weight", "transformer_encoder.layers.2.linear1.bias", "transformer_encoder.layers.2.linear2.weight", "transformer_encoder.layers.2.linear2.bias", "transformer_encoder.layers.2.norm1.weight", "transformer_encoder.layers.2.norm1.bias", "transformer_encoder.layers.2.norm2.weight", "transformer_encoder.layers.2.norm2.bias", "transformer_encoder.layers.2.norm3.weight", "transformer_encoder.layers.2.norm3.bias", "transformer_encoder.layers.3.self_attn.in_proj_weight", "transformer_encoder.layers.3.self_attn.in_proj_bias", "transformer_encoder.layers.3.self_attn.out_proj.weight", "transformer_encoder.layers.3.self_attn.out_proj.bias", "transformer_encoder.layers.3.multihead_attn.in_proj_weight", "transformer_encoder.layers.3.multihead_attn.in_proj_bias", "transformer_encoder.layers.3.multihead_attn.out_proj.weight", "transformer_encoder.layers.3.multihead_attn.out_proj.bias", "transformer_encoder.layers.3.linear1.weight", "transformer_encoder.layers.3.linear1.bias", "transformer_encoder.layers.3.linear2.weight", "transformer_encoder.layers.3.linear2.bias", "transformer_encoder.layers.3.norm1.weight", "transformer_encoder.layers.3.norm1.bias", "transformer_encoder.layers.3.norm2.weight", "transformer_encoder.layers.3.norm2.bias", "transformer_encoder.layers.3.norm3.weight", "transformer_encoder.layers.3.norm3.bias", "transformer_encoder.layers.4.self_attn.in_proj_weight", "transformer_encoder.layers.4.self_attn.in_proj_bias", "transformer_encoder.layers.4.self_attn.out_proj.weight", "transformer_encoder.layers.4.self_attn.out_proj.bias", "transformer_encoder.layers.4.multihead_attn.in_proj_weight", "transformer_encoder.layers.4.multihead_attn.in_proj_bias", "transformer_encoder.layers.4.multihead_attn.out_proj.weight", "transformer_encoder.layers.4.multihead_attn.out_proj.bias", "transformer_encoder.layers.4.linear1.weight", "transformer_encoder.layers.4.linear1.bias", "transformer_encoder.layers.4.linear2.weight", "transformer_encoder.layers.4.linear2.bias", "transformer_encoder.layers.4.norm1.weight", "transformer_encoder.layers.4.norm1.bias", "transformer_encoder.layers.4.norm2.weight", "transformer_encoder.layers.4.norm2.bias", "transformer_encoder.layers.4.norm3.weight", "transformer_encoder.layers.4.norm3.bias", "transformer_encoder.layers.5.self_attn.in_proj_weight", "transformer_encoder.layers.5.self_attn.in_proj_bias", "transformer_encoder.layers.5.self_attn.out_proj.weight", "transformer_encoder.layers.5.self_attn.out_proj.bias", "transformer_encoder.layers.5.multihead_attn.in_proj_weight", "transformer_encoder.layers.5.multihead_attn.in_proj_bias", "transformer_encoder.layers.5.multihead_attn.out_proj.weight", "transformer_encoder.layers.5.multihead_attn.out_proj.bias", "transformer_encoder.layers.5.linear1.weight", "transformer_encoder.layers.5.linear1.bias", "transformer_encoder.layers.5.linear2.weight", "transformer_encoder.layers.5.linear2.bias", "transformer_encoder.layers.5.norm1.weight", "transformer_encoder.layers.5.norm1.bias", "transformer_encoder.layers.5.norm2.weight", "transformer_encoder.layers.5.norm2.bias", "transformer_encoder.layers.5.norm3.weight", "transformer_encoder.layers.5.norm3.bias", "transformer_encoder.norm.weight", "transformer_encoder.norm.bias", "correspondence_decoder.coor_mlp.0.weight", "correspondence_decoder.coor_mlp.0.bias", "correspondence_decoder.coor_mlp.2.weight", "correspondence_decoder.coor_mlp.2.bias", "correspondence_decoder.coor_mlp.4.weight", "correspondence_decoder.coor_mlp.4.bias", "correspondence_decoder.conf_logits_decoder.weight", "correspondence_decoder.conf_logits_decoder.bias", "feature_criterion.W", "feature_criterion_un.W".

what could be possibly the issue here? it seems that backbone or any other block name changed during training and testing? like 'backbone.encoder_blocks....' is missing while 'kpf_encoder.encoder_blocks...' is unexpected keys. Any hint or help would be appreciated!

Hi Yang,

I notice the names of data files in 4DMatch is "camA_xxxx_camB_xxxx.npz". However, there is only one sequence for each shape in DeformingThings4D. So I am wondering what's the meaning of "cam1" and "cam2"? I guess "cam1" is the original camera in DeformingThings4D and "cam2" is generated with a random rigid transformation (fixed for the same shape), is this right?

And I see in the supplementary material the deformation graph is generated according to the geodesic distance computed from the depth images, how can I find the implementation details for this?

Thanks a lot.

Hello! Thanks for open-sourcing this great work, here I got a question about the registration recall reported in the article. From the code : https://github.com/rabbityl/lepard/blob/bcae7f2ee1a2043372f2582b140645f2d6ade9f2/lib/tester.py#L119 It seems that you calculate the pair-level recall instead of the scene-level recall, but in the methods you compared with such as D3Feat or PREDATOR, the metric recall is calculated in scene-level but not in pair-level.

Could you give some hints about it?

Hi, I'd like to ask some questions about the training process,

1. Do you use only one video card during training?
2. How much time does it take to train for an epoch?
3. In configs/train/3Dmatch.yaml, your batch_size = 8, num_worker = 16. I use a 3090 video card. When I use Batch_size = 2 and num_worker =4, I will report an error of insufficient video memory of the video card. Therefore, only batch_size = 1 can be used for training
4. Also, your max_epoch is 1500, so do you need to train for 1500 epochs? Thank you very much for your help.

Hi

I am trying to bring up lepard but I am facing issues while installing C++ extension. Logs are attached below

My system specifications

1. Windows 11
2. using WSL (subsystem for Linux on windows)

Can you please provide some input for solving this?

`running build_ext building 'grid_subsampling' extension Warning: Can't read registry to find the necessary compiler setting Make sure that Python modules winreg, win32api or win32con are installed. C compiler: gcc -pthread -B /home/grblr06/anaconda3/envs/lepard/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC

compile options: '-I/home/grblr06/anaconda3/envs/lepard/lib/python3.8/site-packages/numpy/core/include -I/home/grblr06/anaconda3/envs/lepard/include/python3.8 -c' extra options: '-std=c++11 -D_GLIBCXX_USE_CXX11_ABI=0' gcc: ../cpp_utils/cloud/cloud.cpp gcc: wrapper.cpp gcc: grid_subsampling/grid_subsampling.cpp error: Command "gcc -pthread -B /home/grblr06/anaconda3/envs/lepard/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/home/grblr06/anaconda3/envs/lepard/lib/python3.8/site-packages/numpy/core/include -I/home/grblr06/anaconda3/envs/lepard/include/python3.8 -c grid_subsampling/grid_subsampling.cpp -o build/temp.linux-x86_64-3.8/grid_subsampling/grid_subsampling.o -std=c++11 -D_GLIBCXX_USE_CXX11_ABI=0" failed with exit status 127 running build_ext building 'radius_neighbors' extension Warning: Can't read registry to find the necessary compiler setting Make sure that Python modules winreg, win32api or win32con are installed. C compiler: gcc -pthread -B /home/grblr06/anaconda3/envs/lepard/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC

compile options: '-I/home/grblr06/anaconda3/envs/lepard/lib/python3.8/site-packages/numpy/core/include -I/home/grblr06/anaconda3/envs/lepard/include/python3.8 -c' extra options: '-std=c++11 -D_GLIBCXX_USE_CXX11_ABI=0' gcc: ../cpp_utils/cloud/cloud.cpp gcc: wrapper.cpp gcc: neighbors/neighbors.cpp error: Command "gcc -pthread -B /home/grblr06/anaconda3/envs/lepard/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/home/grblr06/anaconda3/envs/lepard/lib/python3.8/site-packages/numpy/core/include -I/home/grblr06/anaconda3/envs/lepard/include/python3.8 -c neighbors/neighbors.cpp -o build/temp.linux-x86_64-3.8/neighbors/neighbors.o -std=c++11 -D_GLIBCXX_USE_CXX11_ABI=0" failed with exit status 127`

Hi, excellent work! I try to re-train your net with 3dmatch datasets, during the process I found that the parameter 'src_mask' is all True, I think the True or False in 'src_mask' means the point in the source point cloud has the possibility of correspondence to other points in the target point cloud. Am I wrong? or this parameter will change in 4dmatch training process? Hope for your reply!

hi, thank you for sharing such brilliant work! I found the definition of def get_match( conf_matrix, the, mutual=True): is same as the definition of def get_topk_match( conf_matrix, the, mutual=True):. Is there any difference between get_match and get_topk_match? And the paper shows that the authors select matches with confidence higher than a threshold of θc, and further enforce mutual nearest neighbor (MNN) criteria. Does that mean the "get_topk_match" ?

Hi,

Thank you for sharing your excellent work! I have read your paper recently, and I'm wondering what kind of ICP implementaion do you use referred in your paper, and what is Point-to-point ICP.

Hi,

Thank you for sharing your excellent work! I have read your paper recently, and I'm wondering how the draw the point clouds feature map like yours, which I think is really amazing. I will be very appreciative if you can give me some idea about how to draw like this. Thank you very much!

Amazing work! I tried to retrain the dataset, but the loss function did not converge. Have you ever encountered such a situation ? The log file is shown as follows :

#parameters 37.549191 M train Epoch: 1 [ 101/9204]focal_coarse: 3.13 recall_coarse: 0.00 precision_coarse: 0.02 loss: 3.13 train Epoch: 1 [ 202/9204]focal_coarse: 2.97 recall_coarse: 0.00 precision_coarse: 0.02 loss: 2.97 train Epoch: 1 [ 303/9204]focal_coarse: 2.81 recall_coarse: 0.00 precision_coarse: 0.02 loss: 2.81 train Epoch: 1 [ 404/9204]focal_coarse: 2.72 recall_coarse: 0.00 precision_coarse: 0.02 loss: 2.72 train Epoch: 1 [ 505/9204]focal_coarse: 2.65 recall_coarse: 0.00 precision_coarse: 0.02 loss: 2.65 train Epoch: 1 [ 606/9204]focal_coarse: 2.62 recall_coarse: 0.00 precision_coarse: 0.02 loss: 2.62 train Epoch: 1 [ 707/9204]focal_coarse: 2.60 recall_coarse: 0.00 precision_coarse: 0.02 loss: 2.60 train Epoch: 1 [ 808/9204]focal_coarse: 2.56 recall_coarse: 0.00 precision_coarse: 0.02 loss: 2.56 train Epoch: 1 [ 909/9204]focal_coarse: 2.53 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.53 train Epoch: 1 [1010/9204]focal_coarse: 2.50 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.50 train Epoch: 1 [1111/9204]focal_coarse: 2.48 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.48 train Epoch: 1 [1212/9204]focal_coarse: 2.46 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.46 train Epoch: 1 [1313/9204]focal_coarse: 2.44 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.44 train Epoch: 1 [1414/9204]focal_coarse: 2.43 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.43 train Epoch: 1 [1515/9204]focal_coarse: 2.42 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.42 train Epoch: 1 [1616/9204]focal_coarse: 2.41 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.41 train Epoch: 1 [1717/9204]focal_coarse: 2.39 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.39 train Epoch: 1 [1818/9204]focal_coarse: 2.38 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.38 train Epoch: 1 [1919/9204]focal_coarse: 2.37 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.37 train Epoch: 1 [2020/9204]focal_coarse: 2.36 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.36 train Epoch: 1 [2121/9204]focal_coarse: 2.35 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.35 train Epoch: 1 [2222/9204]focal_coarse: 2.34 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.34 train Epoch: 1 [2323/9204]focal_coarse: 2.33 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.33 train Epoch: 1 [2424/9204]focal_coarse: 2.32 recall_coarse: 0.00 precision_coarse: 0.03 loss: 2.32 train Epoch: 1 [2525/9204]focal_coarse: 2.31 recall_coarse: 0.00 precision_coarse: 0.04 loss: 2.31 train Epoch: 1 [2626/9204]focal_coarse: 2.30 recall_coarse: 0.00 precision_coarse: 0.04 loss: 2.30 train Epoch: 1 [2727/9204]focal_coarse: 2.29 recall_coarse: 0.00 precision_coarse: 0.04 loss: 2.29 train Epoch: 1 [2828/9204]focal_coarse: 2.28 recall_coarse: 0.00 precision_coarse: 0.04 loss: 2.28 train Epoch: 1 [2929/9204]focal_coarse: 2.27 recall_coarse: 0.00 precision_coarse: 0.04 loss: 2.27 train Epoch: 1 [3030/9204]focal_coarse: 2.26 recall_coarse: 0.00 precision_coarse: 0.04 loss: 2.26 train Epoch: 1 [3131/9204]focal_coarse: 2.26 recall_coarse: 0.00 precision_coarse: 0.04 loss: 2.26 train Epoch: 1 [3232/9204]focal_coarse: 2.24 recall_coarse: 0.00 precision_coarse: 0.04 loss: 2.24 train Epoch: 1 [3333/9204]focal_coarse: 2.24 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.24 train Epoch: 1 [3434/9204]focal_coarse: 2.23 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.23 train Epoch: 1 [3535/9204]focal_coarse: 2.22 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.22 train Epoch: 1 [3636/9204]focal_coarse: 2.21 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.21 train Epoch: 1 [3737/9204]focal_coarse: 2.20 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.20 train Epoch: 1 [3838/9204]focal_coarse: 2.20 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.20 train Epoch: 1 [3939/9204]focal_coarse: 2.19 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.19 train Epoch: 1 [4040/9204]focal_coarse: 2.18 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.18 train Epoch: 1 [4141/9204]focal_coarse: 2.17 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.17 train Epoch: 1 [4242/9204]focal_coarse: 2.16 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.16 train Epoch: 1 [4343/9204]focal_coarse: 2.15 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.15 train Epoch: 1 [4444/9204]focal_coarse: 2.14 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.14 train Epoch: 1 [4545/9204]focal_coarse: 2.13 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.13 train Epoch: 1 [4646/9204]focal_coarse: 2.13 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.13 train Epoch: 1 [4747/9204]focal_coarse: 2.12 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.12 train Epoch: 1 [4848/9204]focal_coarse: 2.11 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.11 train Epoch: 1 [4949/9204]focal_coarse: 2.10 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.10 train Epoch: 1 [5050/9204]focal_coarse: 2.10 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.10 train Epoch: 1 [5151/9204]focal_coarse: 2.09 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.09 train Epoch: 1 [5252/9204]focal_coarse: 2.08 recall_coarse: 0.00 precision_coarse: 0.08 loss: 2.08 train Epoch: 1 [5353/9204]focal_coarse: 2.07 recall_coarse: 0.00 precision_coarse: 0.08 loss: 2.07 train Epoch: 1 [5454/9204]focal_coarse: 2.06 recall_coarse: 0.00 precision_coarse: 0.08 loss: 2.06 train Epoch: 1 [5555/9204]focal_coarse: 2.05 recall_coarse: 0.00 precision_coarse: 0.08 loss: 2.05 train Epoch: 1 [5656/9204]focal_coarse: 2.05 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.05 train Epoch: 1 [5757/9204]focal_coarse: 2.04 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.04 train Epoch: 1 [5858/9204]focal_coarse: 2.04 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.04 train Epoch: 1 [5959/9204]focal_coarse: 2.03 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.03 train Epoch: 1 [6060/9204]focal_coarse: 2.03 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.03 train Epoch: 1 [6161/9204]focal_coarse: 2.02 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.02 train Epoch: 1 [6262/9204]focal_coarse: 2.02 recall_coarse: 0.00 precision_coarse: 0.10 loss: 2.02 train Epoch: 1 [6363/9204]focal_coarse: 2.01 recall_coarse: 0.00 precision_coarse: 0.10 loss: 2.01 train Epoch: 1 [6464/9204]focal_coarse: 2.01 recall_coarse: 0.00 precision_coarse: 0.10 loss: 2.01 train Epoch: 1 [6565/9204]focal_coarse: 2.00 recall_coarse: 0.00 precision_coarse: 0.10 loss: 2.00 train Epoch: 1 [6666/9204]focal_coarse: 2.00 recall_coarse: 0.00 precision_coarse: 0.10 loss: 2.00 train Epoch: 1 [6767/9204]focal_coarse: 1.99 recall_coarse: 0.00 precision_coarse: 0.10 loss: 1.99 train Epoch: 1 [6868/9204]focal_coarse: 1.99 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.99 train Epoch: 1 [6969/9204]focal_coarse: 1.98 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.98 train Epoch: 1 [7070/9204]focal_coarse: 1.98 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.98 train Epoch: 1 [7171/9204]focal_coarse: 1.97 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.97 train Epoch: 1 [7272/9204]focal_coarse: 1.96 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.96 train Epoch: 1 [7373/9204]focal_coarse: 1.96 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.96 train Epoch: 1 [7474/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [7575/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [7676/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [7777/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [7878/9204]focal_coarse: 1.94 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.94 train Epoch: 1 [7979/9204]focal_coarse: 1.94 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.94 train Epoch: 1 [8080/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [8181/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [8282/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [8383/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [8484/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [8585/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [8686/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [8787/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [8888/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.95 train Epoch: 1 [8989/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.95 train Epoch: 1 [9090/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.95 train Epoch: 1 [9191/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.95 train Epoch: 1focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.95 Save model to experiments/3dmatch-last/2D3D_no_circle_dualmatching/checkpoints/model_1.pth val Epoch: 1focal_coarse: 1.81 recall_coarse: 0.00 precision_coarse: 0.09 loss: 1.81 Save model to experiments/3dmatch-last/2D3D_no_circle_dualmatching/checkpoints/model_best_loss.pth train Epoch: 2 [ 101/9204]focal_coarse: 2.00 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.00 train Epoch: 2 [ 202/9204]focal_coarse: 2.05 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.05 train Epoch: 2 [ 303/9204]focal_coarse: 2.08 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.08 train Epoch: 2 [ 404/9204]focal_coarse: 2.06 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.06 train Epoch: 2 [ 505/9204]focal_coarse: 2.01 recall_coarse: 0.00 precision_coarse: 0.08 loss: 2.01 train Epoch: 2 [ 606/9204]focal_coarse: 1.96 recall_coarse: 0.00 precision_coarse: 0.09 loss: 1.96 train Epoch: 2 [ 707/9204]focal_coarse: 1.94 recall_coarse: 0.00 precision_coarse: 0.10 loss: 1.94 train Epoch: 2 [ 808/9204]focal_coarse: 1.95 recall_coarse: 0.00 precision_coarse: 0.10 loss: 1.95 train Epoch: 2 [ 909/9204]focal_coarse: 1.93 recall_coarse: 0.00 precision_coarse: 0.10 loss: 1.93 train Epoch: 2 [1010/9204]focal_coarse: 1.92 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.92 train Epoch: 2 [1111/9204]focal_coarse: 1.89 recall_coarse: 0.00 precision_coarse: 0.11 loss: 1.89 train Epoch: 2 [1212/9204]focal_coarse: 1.86 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.86 train Epoch: 2 [1313/9204]focal_coarse: 1.86 recall_coarse: 0.00 precision_coarse: 0.12 loss: 1.86 train Epoch: 2 [1414/9204]focal_coarse: 1.84 recall_coarse: 0.00 precision_coarse: 0.13 loss: 1.84 train Epoch: 2 [1515/9204]focal_coarse: 1.82 recall_coarse: 0.00 precision_coarse: 0.14 loss: 1.82 train Epoch: 2 [1616/9204]focal_coarse: 1.81 recall_coarse: 0.00 precision_coarse: 0.14 loss: 1.81 train Epoch: 2 [1717/9204]focal_coarse: 1.81 recall_coarse: 0.00 precision_coarse: 0.14 loss: 1.81 train Epoch: 2 [1818/9204]focal_coarse: 1.80 recall_coarse: 0.00 precision_coarse: 0.15 loss: 1.80 train Epoch: 2 [1919/9204]focal_coarse: 1.81 recall_coarse: 0.00 precision_coarse: 0.14 loss: 1.81 train Epoch: 2 [2020/9204]focal_coarse: 1.81 recall_coarse: 0.00 precision_coarse: 0.15 loss: 1.81 train Epoch: 2 [2121/9204]focal_coarse: 1.85 recall_coarse: 0.00 precision_coarse: 0.14 loss: 1.85 train Epoch: 2 [2222/9204]focal_coarse: 1.85 recall_coarse: 0.00 precision_coarse: 0.14 loss: 1.85 train Epoch: 2 [2323/9204]focal_coarse: 1.92 recall_coarse: 0.00 precision_coarse: 0.14 loss: 1.92 train Epoch: 2 [2424/9204]focal_coarse: 1.98 recall_coarse: 0.00 precision_coarse: 0.13 loss: 1.98 train Epoch: 2 [2525/9204]focal_coarse: 2.04 recall_coarse: 0.00 precision_coarse: 0.13 loss: 2.04 train Epoch: 2 [2626/9204]focal_coarse: 2.09 recall_coarse: 0.00 precision_coarse: 0.12 loss: 2.09 train Epoch: 2 [2727/9204]focal_coarse: 2.14 recall_coarse: 0.00 precision_coarse: 0.12 loss: 2.14 train Epoch: 2 [2828/9204]focal_coarse: 2.19 recall_coarse: 0.00 precision_coarse: 0.11 loss: 2.19 train Epoch: 2 [2929/9204]focal_coarse: 2.23 recall_coarse: 0.00 precision_coarse: 0.11 loss: 2.23 train Epoch: 2 [3030/9204]focal_coarse: 2.27 recall_coarse: 0.00 precision_coarse: 0.11 loss: 2.27 train Epoch: 2 [3131/9204]focal_coarse: 2.31 recall_coarse: 0.00 precision_coarse: 0.10 loss: 2.31 train Epoch: 2 [3232/9204]focal_coarse: 2.35 recall_coarse: 0.00 precision_coarse: 0.10 loss: 2.35 train Epoch: 2 [3333/9204]focal_coarse: 2.38 recall_coarse: 0.00 precision_coarse: 0.10 loss: 2.38 train Epoch: 2 [3434/9204]focal_coarse: 2.41 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.41 train Epoch: 2 [3535/9204]focal_coarse: 2.44 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.44 train Epoch: 2 [3636/9204]focal_coarse: 2.47 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.47 train Epoch: 2 [3737/9204]focal_coarse: 2.49 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.49 train Epoch: 2 [3838/9204]focal_coarse: 2.52 recall_coarse: 0.00 precision_coarse: 0.09 loss: 2.52 train Epoch: 2 [3939/9204]focal_coarse: 2.54 recall_coarse: 0.00 precision_coarse: 0.08 loss: 2.54 train Epoch: 2 [4040/9204]focal_coarse: 2.57 recall_coarse: 0.00 precision_coarse: 0.08 loss: 2.57 train Epoch: 2 [4141/9204]focal_coarse: 2.59 recall_coarse: 0.00 precision_coarse: 0.08 loss: 2.59 train Epoch: 2 [4242/9204]focal_coarse: 2.61 recall_coarse: 0.00 precision_coarse: 0.08 loss: 2.61 train Epoch: 2 [4343/9204]focal_coarse: 2.63 recall_coarse: 0.00 precision_coarse: 0.08 loss: 2.63 train Epoch: 2 [4444/9204]focal_coarse: 2.65 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.65 train Epoch: 2 [4545/9204]focal_coarse: 2.66 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.66 train Epoch: 2 [4646/9204]focal_coarse: 2.68 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.68 train Epoch: 2 [4747/9204]focal_coarse: 2.70 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.70 train Epoch: 2 [4848/9204]focal_coarse: 2.71 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.71 train Epoch: 2 [4949/9204]focal_coarse: 2.73 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.73 train Epoch: 2 [5050/9204]focal_coarse: 2.74 recall_coarse: 0.00 precision_coarse: 0.07 loss: 2.74 train Epoch: 2 [5151/9204]focal_coarse: 2.76 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.76 train Epoch: 2 [5252/9204]focal_coarse: 2.77 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.77 train Epoch: 2 [5353/9204]focal_coarse: 2.78 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.78 train Epoch: 2 [5454/9204]focal_coarse: 2.80 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.80 train Epoch: 2 [5555/9204]focal_coarse: 2.81 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.81 train Epoch: 2 [5656/9204]focal_coarse: 2.82 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.82 train Epoch: 2 [5757/9204]focal_coarse: 2.83 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.83 train Epoch: 2 [5858/9204]focal_coarse: 2.84 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.84 train Epoch: 2 [5959/9204]focal_coarse: 2.85 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.85 train Epoch: 2 [6060/9204]focal_coarse: 2.86 recall_coarse: 0.00 precision_coarse: 0.06 loss: 2.86 train Epoch: 2 [6161/9204]focal_coarse: 2.87 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.87 train Epoch: 2 [6262/9204]focal_coarse: 2.88 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.88 train Epoch: 2 [6363/9204]focal_coarse: 2.89 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.89 train Epoch: 2 [6464/9204]focal_coarse: 2.90 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.90 train Epoch: 2 [6565/9204]focal_coarse: 2.91 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.91 train Epoch: 2 [6666/9204]focal_coarse: 2.91 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.91 train Epoch: 2 [6767/9204]focal_coarse: 2.92 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.92 train Epoch: 2 [6868/9204]focal_coarse: 2.93 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.93 train Epoch: 2 [6969/9204]focal_coarse: 2.94 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.94 train Epoch: 2 [7070/9204]focal_coarse: 2.94 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.94 train Epoch: 2 [7171/9204]focal_coarse: 2.95 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.95 train Epoch: 2 [7272/9204]focal_coarse: 2.96 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.96 train Epoch: 2 [7373/9204]focal_coarse: 2.97 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.97 train Epoch: 2 [7474/9204]focal_coarse: 2.97 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.97 train Epoch: 2 [7575/9204]focal_coarse: 2.98 recall_coarse: 0.00 precision_coarse: 0.05 loss: 2.98 train Epoch: 2 [7676/9204]focal_coarse: 2.98 recall_coarse: 0.00 precision_coarse: 0.04 loss: 2.98 train Epoch: 2 [7777/9204]focal_coarse: 2.99 recall_coarse: 0.00 precision_coarse: 0.04 loss: 2.99 train Epoch: 2 [7878/9204]focal_coarse: 3.00 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.00 train Epoch: 2 [7979/9204]focal_coarse: 3.00 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.00 train Epoch: 2 [8080/9204]focal_coarse: 3.01 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.01 train Epoch: 2 [8181/9204]focal_coarse: 3.01 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.01 train Epoch: 2 [8282/9204]focal_coarse: 3.02 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.02 train Epoch: 2 [8383/9204]focal_coarse: 3.02 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.02 train Epoch: 2 [8484/9204]focal_coarse: 3.03 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.03 train Epoch: 2 [8585/9204]focal_coarse: 3.03 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.03 train Epoch: 2 [8686/9204]focal_coarse: 3.04 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.04 train Epoch: 2 [8787/9204]focal_coarse: 3.04 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.04 train Epoch: 2 [8888/9204]focal_coarse: 3.05 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.05 train Epoch: 2 [8989/9204]focal_coarse: 3.05 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.05 train Epoch: 2 [9090/9204]focal_coarse: 3.06 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.06 train Epoch: 2 [9191/9204]focal_coarse: 3.06 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.06 train Epoch: 2focal_coarse: 3.06 recall_coarse: 0.00 precision_coarse: 0.04 loss: 3.06 Save model to experiments/3dmatch-last/2D3D_no_circle_dualmatching/checkpoints/model_2.pth val Epoch: 2focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [ 101/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 3 [ 202/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 3 [ 303/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [ 404/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [ 505/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.44 train Epoch: 3 [ 606/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.44 train Epoch: 3 [ 707/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.44 train Epoch: 3 [ 808/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [ 909/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [1010/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [1111/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.44 train Epoch: 3 [1212/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [1313/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [1414/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [1515/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [1616/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [1717/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [1818/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [1919/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [2020/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [2121/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [2222/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [2323/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [2424/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [2525/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [2626/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [2727/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [2828/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [2929/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [3030/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [3131/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [3232/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [3333/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [3434/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [3535/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [3636/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [3737/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [3838/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [3939/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [4040/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [4141/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [4242/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [4343/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [4444/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [4545/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [4646/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [4747/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [4848/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [4949/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [5050/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [5151/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [5252/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [5353/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [5454/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 3 [5555/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [5656/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [5757/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [5858/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [5959/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [6060/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [6161/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [6262/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [6363/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [6464/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [6565/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [6666/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [6767/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [6868/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [6969/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [7070/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [7171/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [7272/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [7373/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [7474/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [7575/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [7676/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [7777/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [7878/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [7979/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [8080/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [8181/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [8282/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [8383/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [8484/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [8585/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [8686/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [8787/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [8888/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [8989/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [9090/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3 [9191/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 3focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 Save model to experiments/3dmatch-last/2D3D_no_circle_dualmatching/checkpoints/model_3.pth val Epoch: 3focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [ 101/9204]focal_coarse: 3.42 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.42 train Epoch: 4 [ 202/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [ 303/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [ 404/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [ 505/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [ 606/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [ 707/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [ 808/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [ 909/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [1010/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [1111/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [1212/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [1313/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [1414/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [1515/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.44 train Epoch: 4 [1616/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [1717/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [1818/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [1919/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [2020/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [2121/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [2222/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [2323/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [2424/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [2525/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [2626/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [2727/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [2828/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [2929/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [3030/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [3131/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [3232/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [3333/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [3434/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [3535/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [3636/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [3737/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [3838/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [3939/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [4040/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [4141/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [4242/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [4343/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [4444/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [4545/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [4646/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [4747/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [4848/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [4949/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [5050/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [5151/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [5252/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [5353/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [5454/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [5555/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [5656/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [5757/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [5858/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [5959/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [6060/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [6161/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [6262/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [6363/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [6464/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [6565/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [6666/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [6767/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 4 [6868/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [6969/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [7070/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [7171/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [7272/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [7373/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [7474/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [7575/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [7676/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [7777/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [7878/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [7979/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [8080/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [8181/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [8282/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [8383/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [8484/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [8585/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [8686/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [8787/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [8888/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [8989/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [9090/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4 [9191/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 4focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 Save model to experiments/3dmatch-last/2D3D_no_circle_dualmatching/checkpoints/model_4.pth val Epoch: 4focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [ 101/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [ 202/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [ 303/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [ 404/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [ 505/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [ 606/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [ 707/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [ 808/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [ 909/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [1010/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [1111/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [1212/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [1313/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [1414/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [1515/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [1616/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [1717/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [1818/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [1919/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 5 [2020/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 5 [2121/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 5 [2222/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 5 [2323/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 5 [2424/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 5 [2525/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [2626/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [2727/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [2828/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [2929/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 5 [3030/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 5 [3131/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [3232/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 5 [3333/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 5 [3434/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 5 [3535/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 5 [3636/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 5 [3737/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 5 [3838/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 5 [3939/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [4040/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [4141/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [4242/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [4343/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [4444/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [4545/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [4646/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [4747/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [4848/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [4949/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [5050/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [5151/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [5252/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [5353/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [5454/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [5555/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [5656/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [5757/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [5858/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [5959/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [6060/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [6161/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [6262/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [6363/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [6464/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [6565/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [6666/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [6767/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [6868/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [6969/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [7070/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [7171/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [7272/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [7373/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [7474/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [7575/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [7676/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [7777/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [7878/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [7979/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [8080/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [8181/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [8282/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [8383/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [8484/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [8585/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [8686/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [8787/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [8888/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [8989/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [9090/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5 [9191/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 5focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 Save model to experiments/3dmatch-last/2D3D_no_circle_dualmatching/checkpoints/model_5.pth val Epoch: 5focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [ 101/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [ 202/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [ 303/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [ 404/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [ 505/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [ 606/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [ 707/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [ 808/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [ 909/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [1010/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [1111/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [1212/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [1313/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [1414/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [1515/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [1616/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [1717/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [1818/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [1919/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [2020/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [2121/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.44 train Epoch: 6 [2222/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 6 [2323/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.44 train Epoch: 6 [2424/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [2525/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [2626/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.44 train Epoch: 6 [2727/9204]focal_coarse: 3.44 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.44 train Epoch: 6 [2828/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [2929/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [3030/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [3131/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [3232/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.01 loss: 3.45 train Epoch: 6 [3333/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [3434/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [3535/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [3636/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [3737/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [3838/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [3939/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [4040/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [4141/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [4242/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45 train Epoch: 6 [4343/9204]focal_coarse: 3.45 recall_coarse: 0.00 precision_coarse: 0.00 loss: 3.45