Unable to train on the Replica_Edge dataset

[C-H-Chien]

Hi! I met an issue when trying to train on the Replica_Edge dataset. It throws the error message when executing bash scripts/run_Replica.bash as suggested by README,

(emap) $ bash scripts/run_Replica.bash [main.py:77 - main() ] Run on scan of room0 Load data: Begin Load data: End [main.py:86 - main() ] Extracting edges from UDF [runner_udf.py:514 - extract_edge() ] Find checkpoint: ckpt_best.pth Traceback (most recent call last): File "main.py", line 102, in main() File "main.py", line 87, in main runner.extract_edge( File "/EMAP/src/runner/runner_udf.py", line 515, in extract_edge self.load_checkpoint(latest_model_name) File "/EMAP/src/runner/runner_udf.py", line 253, in load_checkpoint checkpoint = torch.load( File "/.conda/envs/emap/lib/python3.8/site-packages/torch/serialization.py", line 986, in load with _open_file_like(f, 'rb') as opened_file: File "/.conda/envs/emap/lib/python3.8/site-packages/torch/serialization.py", line 435, in _open_file_like return _open_file(name_or_buffer, mode) File "/.conda/envs/emap/lib/python3.8/site-packages/torch/serialization.py", line 416, in init super().init(open(name, mode)) FileNotFoundError: [Errno 2] No such file or directory: './exp/Replica/room0/emap/checkpoints/ckpt_best.pth'

Training on ABC-NEF and DTU_Edge datasets do not have such an issue. Is there something I did wrongly? Thank you!

[rayeeli]

Hi, the code for training the UDF field was mistakenly commented out. It should be working now.

[C-H-Chien]

Thank you for the updates! That resolves the issue. Closing this thread.