hturki/suds

Issues

• training/dino_02/0000/000000.parquet.pt': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.

  #49 opened 4 months ago by derrick-xwp
  0

• No module named 'suds.data.image_metadata'

  #48 opened 4 months ago by derrick-xwp
  0

• ModuleNotFoundError: No module named 'scripts.train'

  #47 opened 4 months ago by derrick-xwp
  0

• eval

  #46 opened 9 months ago by baekhyun77
  0

• How many frames did you take for each sequence?

  #39 opened a year ago by jelleopard
  1

• Static and dynamic decomposing quality

  #45 opened a year ago by ntaquan0125
  0

• conda errors

  #44 opened a year ago by antithing
  0

• export pointcloud error: AttributeError: 'SUDSDataManager' object has no attribute 'train_pixel_sampler

  #37 opened a year ago by heatMa
  1

• unexpected keyword argument

  #43 opened a year ago by baekhyun77
  0

• RuntimeError: Encoding 'SequentialGrid' not found

  #36 opened a year ago by vermouth599
  4

• import error

  #42 opened a year ago by baekhyun77
  0

• out of memory error

  #41 opened a year ago by rajatmodi62
  0

• Spatial alignment and appearance gap

  #40 opened a year ago by Beastmaster
  0

• Static-dynamic disentanglement

  #38 opened a year ago by samiragarwala
  0

• RuntimeError: Encoding 'SequentialGrid' not found

  #15 opened 2 years ago by tb2-sy
  1

• Baselines

  #35 opened a year ago by samiragarwala
  2

• Adjust Batch Size

  #34 opened a year ago by Tao-11-chen
  0

• Process finished with exit code 137 (interrupted by signal 9: SIGKILL)

  #32 opened 2 years ago by zzyycc7
  4

• What's the different of "DynamicFiled" and "DynamicProposalField"?

  #33 opened 2 years ago by rockywind
  0

• Why the depth loss flucture so much?

  #30 opened 2 years ago by rockywind
  2

• How to add videos from different scenarios into the same JSON file for training?

  #31 opened 2 years ago by rockywind
  2

• Missing files

  #29 opened 2 years ago by OrangeSodahub
  5

• Question about the discrepancy between the paper's description and the code implementation.

  #25 opened 2 years ago by rockywind
  1

• scripts/run_pca.py uses excessive memory

  #26 opened 2 years ago by dllu
  1

• Potentially redundant for-loop in extract_dino_correspondences.py

  #27 opened 2 years ago by dllu
  1

• "in human_format [17243] magnitude = int(floor(log(num, k))) [17244] ValueError: math domain error"

  #24 opened 2 years ago by rockywind
  0

• How do you decompose dynamic objects and backgrounds?

  #19 opened 2 years ago by shanxiaojun
  3

• Do you consider Nuscenes/waymo support?

  #23 opened 2 years ago by sxj1215
  0

• Why use .parquet for depth and optical flow？

  #21 opened 2 years ago by AIBUWAN
  1

• How to get the world-scale depth

  #22 opened 2 years ago by rockywind
  3

• How do you eval your model?

  #20 opened 2 years ago by shanxiaojun
  1

• question about dynamic RGB

  #17 opened 2 years ago by tb2-sy
  1

• Could not find a version that satisfies the requirement

  #3 opened 2 years ago by shanxiaojun
  9

• How to generate the semantic label?

  #18 opened 2 years ago by rockywind
  1

• How to generate the demo video in the project page including ego-vehicle shifting, panoptical segmentation and shorting the camera focal length.

  #13 opened 2 years ago by rockywind
  3

• multi-GPU Training

  #16 opened 2 years ago by tb2-sy
  2

• How to set the "FEATURE_CLUSTERS" in the training script?

  #4 opened 2 years ago by rockywind
  5

• How to generate the .json files in the metadata folder?

  #14 opened 2 years ago by JaamieZheng
  3

• Is the tiny-cuda-nn updated?

  #1 opened 2 years ago by h8c2
  5

• Compute correspondence too slow!

  #9 opened 2 years ago by rockywind
  7

• How to deal with kitti odom?

  #12 opened 2 years ago by AIBUWAN
  1

• Do you consider add nuscenes support?

  #10 opened 2 years ago by synsin0
  2

• city-1M

  #8 opened 2 years ago by tb2-sy
  1

• "RuntimeError: CUDA error: device-side assert triggered CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect."

  #5 opened 2 years ago by rockywind
  1

• Downsample rate before feature extraction

  #7 opened 2 years ago by rwn17
  2

• " return _C.create_encoding(self.n_input_dims, self.encoding_config, self.precision) RuntimeError: Invalid encoding type: SequentialGrid"

  #6 opened 2 years ago by rockywind
  2

• “No module named 'suds'”

  #2 opened 2 years ago by rockywind
  1