-
-
Notifications
You must be signed in to change notification settings - Fork 1.8k
Open
Labels
bugSomething isn't workingSomething isn't workinghowtousedlcissue regarding the use of DeepLabCutissue regarding the use of DeepLabCut
Description
Is there an existing issue for this?
- I have searched the existing issues
Operating System
Tahoe 26.2 (25C56)
DeepLabCut version
3.0.0rc13
What engine are you using?
pytorch
DeepLabCut mode
multi animal
Device type
M2 Max
Bug description 🐛
When i try to create a dataset (after labeling 50 frames), with the following training settings:
I cannot create a training dataset.
If I use resnet_50 as a model (settings below), I do NOT get this problem.
See logs for error messages
Steps To Reproduce
Under Create Training dataset:
choose Transfer Learning superanimal -- topviewmouse for weight initialization. The rest of the fields get set automatically.
The labeling part does not seem to matter as long as it's valid
click "Create Training Dataset"
Dataset creation silently fails (no pop-up message, only in terminal output)
Relevant log output
Loading DLC 3.0.0rc13...
INFO:console:Training dataset creation failed.
superanimal_topviewmouse_hrnet_w32.pt: 100%|███████████████████████████████████████████████████| 118M/118M [00:02<00:00, 52.8MB/s]
Traceback (most recent call last):
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/tarfile.py", line 1907, in gzopen
t = cls.taropen(name, mode, fileobj, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/tarfile.py", line 1884, in taropen
return cls(name, mode, fileobj, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/tarfile.py", line 1744, in __init__
self.firstmember = self.next()
^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/tarfile.py", line 2662, in next
raise e
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/tarfile.py", line 2635, in next
tarinfo = self.tarinfo.fromtarfile(self)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/tarfile.py", line 1296, in fromtarfile
buf = tarfile.fileobj.read(BLOCKSIZE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/gzip.py", line 301, in read
return self._buffer.read(size)
^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/_compression.py", line 68, in readinto
data = self.read(len(byte_view))
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/gzip.py", line 499, in read
if not self._read_gzip_header():
^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/gzip.py", line 468, in _read_gzip_header
last_mtime = _read_gzip_header(self._fp)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/gzip.py", line 428, in _read_gzip_header
raise BadGzipFile('Not a gzipped file (%r)' % magic)
gzip.BadGzipFile: Not a gzipped file (b'PK')
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/site-packages/dlclibrary/dlcmodelzoo/modelzoo_download.py", line 120, in _handle_downloaded_file
with tarfile.open(file_path, mode="r:gz") as tar:
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/tarfile.py", line 1854, in open
return func(name, filemode, fileobj, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/tarfile.py", line 1911, in gzopen
raise ReadError("not a gzip file") from e
tarfile.ReadError: not a gzip file
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/site-packages/deeplabcut/gui/tabs/create_training_dataset.py", line 278, in create_training_dataset
self.weight_init_selector.get_super_animal_weight_init(
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/site-packages/deeplabcut/gui/tabs/create_training_dataset.py", line 710, in get_super_animal_weight_init
weight_init = build_weight_init(
^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/site-packages/deeplabcut/modelzoo/weight_initialization.py", line 91, in build_weight_init
snapshot_path = get_super_animal_snapshot_path(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/site-packages/deeplabcut/pose_estimation_pytorch/modelzoo/utils.py", line 83, in get_super_animal_snapshot_path
download_super_animal_snapshot(dataset, model_name)
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/site-packages/deeplabcut/pose_estimation_pytorch/modelzoo/utils.py", line 146, in download_super_animal_snapshot
download_huggingface_model(
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/site-packages/dlclibrary/dlcmodelzoo/modelzoo_download.py", line 183, in download_huggingface_model
_handle_downloaded_file(file_name, target_dir, rename_mapping)
File "/Users/mike/.conda/envs/dlc3/lib/python3.11/site-packages/dlclibrary/dlcmodelzoo/modelzoo_download.py", line 127, in _handle_downloaded_file
file_name = rename_mapping.get(file_name, file_name)
^^^^^^^^^^^^^^^^^^
AttributeError: 'str' object has no attribute 'get'
INFO:console:Loading....
INFO:console:
INFO:console:superanimal_topviewmouse_hrnet_w32Anything else?
No response
Code of Conduct
- I agree to follow this project's Code of Conduct
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
bugSomething isn't workingSomething isn't workinghowtousedlcissue regarding the use of DeepLabCutissue regarding the use of DeepLabCut