Hypervisor error (0x20001) when running training

[sophiaanner]

I've been having several problems since upgrading to the newest version of SLEAP (#2417 and #2442) and I'm hoping the SLEAP community can help me out with this one too.

I can open my SLEAP project, make predictions, etc., but when I go to run training, my computer will crash and give me the following message:

and here's what the command prompt looks like right before that:

I've tried nearly every solution I can find online, with attempts to run training in SLEAP after each one, including:

• updating Windows
• updating drivers
• uninstalling and reinstalling SLEAP
• moving my videos to my desktop so I don't have any external devices plugged in
• checking my drivers for error
• using the windows memory diagnostic tool (no errors)
• enabling the hypervisor platform and virtual machine platform in "Turn Windows features on or off"
• disable hypervisor via command prompt as administrator
• disable virtualization support in BIOS advanced settings

I was able to run training one time on v1.5.2 before running into this problem.

Admittedly I don't know very much about computers, but this laptop is brand new so shouldn't have too many issues and I only need/use it for SLEAP. Here's some more relevant information:

OS: Windows 11 Pro
Version(s): SLEAP 1.5.1; python 3.11
SLEAP installation method: uv tool install (python 3.13)
GPU: NVIDIA RTX PRO 500 Blackwell Generation Laptop GPU (CUDA 12.8)

Has anyone else run into this error or know at all what might be going on? My next step may be to fully reset my computer and Windows.

Much appreciated,
Sophia

[talmo]

Hi @sophiaanner,

Sorry about the issues! This seems to be a system one rather than on the SLEAP side, but let's see what we can do.

1. I know you tried updating drivers, but did you try specifically updating the GPU drivers from the NVIDIA drivers website as opposed to Windows update? It looks like R580 U4 (581.42) is the latest version FWIW.
2. It sounds like you did try disabling the hypervisor stuff, but just to be sure, you can try disabling all of these in the "Turn Windows features on or off" settings: Hyper-V, Windows Hypervisor Platform, Virtual Machine Platform, WSL, and Core Isolation → Memory Integrity.
3. One more to try is: NVIDIA Control Panel → Manage 3D settings → Power management mode = Prefer maximum performance

This is a weird one, but unfortunately GPU laptops are a bit weird and the Blackwell ones are especially bleeding edge.

Can you also check the versions installed in the tool environment? In the Command Prompt:

for /f %i in ('uv tool dir') do uv pip list --python %i/sleap/Scripts/python.exe

You can just paste all the versions here. One thing we want to make sure is that it's using PyTorch 2.8 for all the deps. Should be the case, but let's be sure :)

Cheers,

Talmo

[sophiaanner]

Hi Talmo,

Thanks for the quick response. I update my driver from the NVIDIA website to 581.42 as you recommended and was able to begin training without the computer crashing! Unfortunately, I get an error message that the centroid model was unable to run. The centered_instance training did run, and the command line says inference was completed but no predictions showed up on my video.

I uninstalled and reinstalled SLEAP after updating my NVIDIA driver, and noticed that when I run sleap-diagnostic, the CUDA for this driver is now 13.0. Could that be part of the problem?

Here is the extensive error message related to not being able to run the centroid model:
Sanity Checking DataLoader 0: 0%| | 0/2 [00:00<?, ?it/s]2025-11-04 11:23:31 | INFO | sleap_nn.training.model_trainer:train:1037 | Finished training loop. [0.0 min]
Traceback (most recent call last):
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\call.py", line 49, in _call_and_handle_interrupt
return trainer_fn(*args, **kwargs)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\trainer.py", line 598, in _fit_impl
self._run(model, ckpt_path=ckpt_path)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\trainer.py", line 1011, in _run
results = self._run_stage()
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\trainer.py", line 1053, in _run_stage
self._run_sanity_check()
~~~~~~~~~~~~~~~~~~~~~~^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\trainer.py", line 1082, in _run_sanity_check
val_loop.run()
~~~~~~~~~~~~^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\loops\utilities.py", line 179, in _decorator
return loop_run(self, *args, **kwargs)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\loops\evaluation_loop.py", line 145, in run
self._evaluation_step(batch, batch_idx, dataloader_idx, dataloader_iter)
~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\loops\evaluation_loop.py", line 437, in _evaluation_step
output = call._call_strategy_hook(trainer, hook_name, *step_args)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\call.py", line 329, in _call_strategy_hook
output = fn(*args, **kwargs)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\strategies\strategy.py", line 412, in validation_step
return self.lightning_module.validation_step(*args, **kwargs)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_nn\training\lightning_modules.py", line 973, in validation_step
y_preds = self.model(X)["CentroidConfmapsHead"]
~~~~~~~~~~^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1773, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1784, in _call_impl
return forward_call(*args, **kwargs)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_nn\architectures\model.py", line 195, in forward
outputs[head.name] = head_layer(backbone_outputs["outputs"][idx])
~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1773, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1784, in _call_impl
return forward_call(*args, **kwargs)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\container.py", line 244, in forward
input = module(input)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1773, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1784, in _call_impl
return forward_call(*args, **kwargs)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\container.py", line 244, in forward
input = module(input)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1773, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1784, in _call_impl
return forward_call(*args, **kwargs)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\conv.py", line 548, in forward
return self._conv_forward(input, self.weight, self.bias)
~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\conv.py", line 543, in _conv_forward
return F.conv2d(
~~~~~~~~^
input, weight, bias, self.stride, self.padding, self.dilation, self.groups
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
RuntimeError: GET was unable to find an engine to execute this computation

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "", line 198, in _run_module_as_main
File "", line 88, in run_code
File "C:\Users\sophi.local\bin\sleap-nn-train.exe_main.py", line 10, in
sys.exit(train())
~~~~~^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\click\core.py", line 1462, in call
return self.main(*args, **kwargs)
~~~~~~~~~^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\click\core.py", line 1383, in main
rv = self.invoke(ctx)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\click\core.py", line 1246, in invoke
return ctx.invoke(self.callback, **ctx.params)
~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\click\core.py", line 814, in invoke
return callback(*args, **kwargs)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\cli.py", line 87, in train
run_training(cfg)
~~~~~~~~~~~~^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_nn\train.py", line 30, in run_training
trainer.train()
~~~~~~~~~~~~~^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_nn\training\model_trainer.py", line 1026, in train
self.trainer.fit(
~~~~~~~~~~~~~~~~^
self.lightning_model,
^^^^^^^^^^^^^^^^^^^^^
...<2 lines>...
ckpt_path=self.config.trainer_config.resume_ckpt_path,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\trainer.py", line 560, in fit
call._call_and_handle_interrupt(
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^
self, self._fit_impl, model, train_dataloaders, val_dataloaders, datamodule, ckpt_path
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\call.py", line 70, in _call_and_handle_interrupt
trainer._teardown()
~~~~~~~~~~~~~~~~~^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\trainer.py", line 1034, in _teardown
self.strategy.teardown()
~~~~~~~~~~~~~~~~~~~~~~^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\strategies\strategy.py", line 536, in teardown
self.lightning_module.cpu()
~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\fabric\utilities\device_dtype_mixin.py", line 82, in cpu
return super().cpu()
~~~~~~~~~~~^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1150, in cpu
return self._apply(lambda t: t.cpu())
~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 928, in _apply
module._apply(fn)
~~~~~~~~~~~~~^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 928, in _apply
module._apply(fn)
~~~~~~~~~~~~~^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 928, in _apply
module._apply(fn)
~~~~~~~~~~~~~^^^^
[Previous line repeated 5 more times]
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 955, in _apply
param_applied = fn(param)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1150, in
return self._apply(lambda t: t.cpu())
~~~~~^^
torch.AcceleratorError: CUDA error: an illegal memory access was encountered
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

Unhandled exception caught in c10/util/AbortHandler.h
00007FFD16296A0400007FFD1628E160 torch_python.dll!torch::autograd::THPCppFunction_requires_grad [ @ ]
00007FFEF68D19D700007FFEF68D19C0 ucrtbase.dll!terminate [ @ ]
00007FFEF1DC1943 VCRUNTIME140_1.dll! [ @ ]
00007FFEF1DC21AF VCRUNTIME140_1.dll! [ @ ]
00007FFEF1DC220D VCRUNTIME140_1.dll! [ @ ]
00007FFEF1DC408200007FFEF1DC3FD0 VCRUNTIME140_1.dll!_CxxFrameHandler4 [ @ ]
00007FFEF95C632F00007FFEF95C6290 ntdll.dll!_chkstk [ @ ]
00007FFEF947232700007FFEF9471D90 ntdll.dll!RtlLocateExtendedFeature [ @ ]
00007FFEF946A96100007FFEF946A740 ntdll.dll!RtlRaiseException [ @ ]
00007FFEF6E4804A00007FFEF6E47FC0 KERNELBASE.dll!RaiseException [ @ ]
00007FFEF2BE52C700007FFEF2BE5230 VCRUNTIME140.dll!CxxThrowException [ @ ]
00007FFE802381A600007FFE80238050 c10_cuda.dll!c10::cuda::c10_cuda_check_implementation [ @ ]
00007FFD285CBFB000007FFD285CBF40 torch_cuda.dll!at::cuda::CUDAEvent::createEvent [ @ ]
00007FFD285CF1FE00007FFD285CF1C0 torch_cuda.dll!at::cuda::CUDAEvent::record [ @ ]
00007FFD285DEBE500007FFD285DB280 torch_cuda.dll!at::cuda::flush_icache [ @ ]
00007FFD285DCCF900007FFD285DB280 torch_cuda.dll!at::cuda::flush_icache [ @ ]
00007FFEB46F495E00007FFEB46F4910 c10.dll!c10::StorageImpl::~StorageImpl [ @ ]
00007FFEB46F560700007FFEB46F5450 c10.dll!c10::SymBool::operator|| [ @ ]
00007FFEB46DFB9200007FFEB46DFB20 c10.dll!c10::ConstantSymNodeImpl::~ConstantSymNodeImpl [ @ ]
00007FFEB471927C00007FFEB4719220 c10.dll!c10::TensorImpl::~TensorImpl [ @ ]
00007FFD1CD6438500007FFD1CD62B70 torch_cpu.dll!at::DynamicLibrary::sym [ @ ]
00007FFD1CD096C800007FFD1CD09640 torch_cpu.dll!at::TensorBase::reset [ @ ]
00007FFD162A584500007FFD162972B0 torch_python.dll!initModule [ @ ]
00007FFD1634B5A500007FFD16349120 torch_python.dll!THPVariable_Wrap [ @ ]
00007FFD1634B8B300007FFD16349120 torch_python.dll!THPVariable_Wrap [ @ ]
00007FFE7B7FF72400007FFE7B7FD68C python313.dll!PyObject_Malloc [ @ ]
00007FFE7B769B0300007FFE7B769884 python313.dll!PyObject_GC_UnTrack [ @ ]
00007FFE7B76974400007FFE7B768E20 python313.dll!PyThreadState_Get [ @ ]
00007FFE7B76973A00007FFE7B768E20 python313.dll!PyThreadState_Get [ @ ]
00007FFE7B76973A00007FFE7B768E20 python313.dll!PyThreadState_Get [ @ ]
00007FFE7B76973A00007FFE7B768E20 python313.dll!PyThreadState_Get [ @ ]
00007FFE7B76973A00007FFE7B768E20 python313.dll!PyThreadState_Get [ @ ]
00007FFE7B76973A00007FFE7B768E20 python313.dll!PyThreadState_Get [ @ ]
00007FFE7B76973A00007FFE7B768E20 python313.dll!PyThreadState_Get [ @ ]
00007FFE7B813C1800007FFE7B813670 python313.dll!PyUnicodeWriter_Init [ @ ]
00007FFE7B813C2D00007FFE7B813670 python313.dll!PyUnicodeWriter_Init [ @ ]
00007FFE7B7D67FA00007FFE7B7D6208 python313.dll!PyObject_CallOneArg [ @ ]
00007FFE7B7B488D00007FFE7B7B4430 python313.dll!PyDict_SetItem_Take2 [ @ ]
00007FFE7B85F42F00007FFE7B85F37C python313.dll!PyDict_SetItemString [ @ ]
00007FFE7B85756E00007FFE7B856EB8 python313.dll!Py_GetBuildInfo [ @ ]
00007FFE7B85747700007FFE7B856EB8 python313.dll!Py_GetBuildInfo [ @ ]
00007FFE7B819E5D00007FFE7B819494 python313.dll!PyObject_Type [ @ ]
00007FFE7B8741CF00007FFE7B873A28 python313.dll!PyThread_tss_delete [ @ ]
00007FFE7B87305D00007FFE7B873040 python313.dll!Py_RunMain [ @ ]
00007FFE7B87283B00007FFE7B872810 python313.dll!Py_Main [ @ ]
00007FF7CC391230 python.exe! [ @ ]
00007FFEF826E8D700007FFEF826E8C0 KERNEL32.DLL!BaseThreadInitThunk [ @ ]
00007FFEF94EC53C00007FFEF94EC510 ntdll.dll!RtlUserThreadStart [ @ ]

And the versions of all packages I have installed:
Package Version

---

aiohappyeyeballs 2.6.1
aiohttp 3.13.2
aiosignal 1.4.0
albucore 0.0.24
albumentations 2.0.8
annotated-types 0.7.0
antlr4-python3-runtime 4.9.3
anyio 4.11.0
argon2-cffi 25.1.0
argon2-cffi-bindings 25.1.0
arrow 1.4.0
asttokens 3.0.0
async-lru 2.0.5
attrs 25.4.0
av 15.1.0
babel 2.17.0
beautifulsoup4 4.14.2
bleach 6.3.0
certifi 2022.12.7
cffi 2.0.0
charset-normalizer 2.1.1
click 8.3.0
colorama 0.4.6
comm 0.2.3
contourpy 1.3.3
cycler 0.12.1
debugpy 1.8.17
decorator 5.2.1
defusedxml 0.7.1
executing 2.2.1
fastjsonschema 2.21.2
filelock 3.19.1
fonttools 4.60.1
fqdn 1.5.1
frozenlist 1.8.0
fsspec 2025.9.0
gitdb 4.0.12
gitpython 3.1.45
h11 0.16.0
h5py 3.15.1
hdmf 4.1.1
httpcore 1.0.9
httpx 0.28.1
hydra-core 1.3.2
idna 3.4
imageio 2.37.2
imageio-ffmpeg 0.6.0
imgstore 0.3.7
ipykernel 7.1.0
ipython 9.6.0
ipython-pygments-lexers 1.1.1
ipywidgets 8.1.8
isoduration 20.11.0
jedi 0.19.2
jinja2 3.1.6
joblib 1.5.2
jsmin 3.0.1
json5 0.12.1
jsonpickle 4.1.1
jsonpointer 3.0.0
jsonschema 4.25.1
jsonschema-specifications 2025.9.1
jupyter 1.1.1
jupyter-client 8.6.3
jupyter-console 6.6.3
jupyter-core 5.9.1
jupyter-events 0.12.0
jupyter-lsp 2.3.0
jupyter-server 2.17.0
jupyter-server-terminals 0.5.3
jupyterlab 4.1.6
jupyterlab-pygments 0.3.0
jupyterlab-server 2.24.0
jupyterlab-widgets 3.0.16
kiwisolver 1.4.9
kornia 0.8.1
kornia-rs 0.1.9
lark 1.3.1
lazy-loader 0.4
lightning 2.5.5
lightning-utilities 0.11.8
loguru 0.7.3
markdown-it-py 4.0.0
markupsafe 2.1.5
matplotlib 3.10.7
matplotlib-inline 0.2.1
mdurl 0.1.2
mistune 3.1.4
mpmath 1.3.0
multidict 6.7.0
nbclient 0.10.2
nbconvert 7.16.6
nbformat 5.10.4
ndx-pose 0.2.2
nest-asyncio 1.6.0
networkx 3.5
nixio 1.5.4
notebook 7.1.3
notebook-shim 0.2.4
numpy 2.3.3
omegaconf 2.3.0
opencv-python 4.11.0.86
opencv-python-headless 4.11.0.86
packaging 24.1
pandas 2.3.3
pandocfilters 1.5.1
parso 0.8.5
pillow 11.3.0
platformdirs 4.5.0
prometheus-client 0.23.1
prompt-toolkit 3.0.52
propcache 0.4.1
protobuf 6.33.0
psutil 7.1.3
pure-eval 0.2.3
pycparser 2.23
pydantic 2.12.3
pydantic-core 2.41.4
pygments 2.19.2
pykalman 0.10.2
pymatreader 1.1.0
pynwb 3.1.2
pyparsing 3.2.5
pyside6 6.10.0
pyside6-addons 6.10.0
pyside6-essentials 6.10.0
python-dateutil 2.9.0.post0
python-json-logger 4.0.0
python-rapidjson 1.22
pytorch-lightning 2.5.5
pytz 2025.2
pywinpty 3.0.2
pyyaml 6.0.3
pyzmq 27.1.0
qtpy 2.4.3
referencing 0.37.0
requests 2.28.1
rfc3339-validator 0.1.4
rfc3986-validator 0.1.1
rfc3987-syntax 1.1.0
rich 14.2.0
rich-click 1.9.4
rpds-py 0.28.0
ruamel-yaml 0.18.16
ruamel-yaml-clib 0.2.14
scikit-base 0.12.6
scikit-image 0.25.2
scikit-learn 1.7.2
scipy 1.16.3
seaborn 0.13.2
send2trash 1.8.3
sentry-sdk 2.43.0
setuptools 70.2.0
shiboken6 6.10.0
simplejson 3.20.2
simsimd 6.5.3
six 1.17.0
sleap 1.5.2
sleap-io 0.5.7
sleap-nn 0.0.4
smmap 5.0.2
sniffio 1.3.1
soupsieve 2.8
stack-data 0.6.3
stringzilla 4.2.3
sympy 1.14.0
terminado 0.18.1
threadpoolctl 3.6.0
tifffile 2025.10.16
tinycss2 1.4.0
torch 2.8.0+cu128
torchmetrics 1.0.3
torchvision 0.23.0+cu128
tornado 6.5.2
tqdm 4.66.5
traitlets 5.14.3
typing-extensions 4.15.0
typing-inspection 0.4.2
tzdata 2025.2
tzlocal 5.3.1
uri-template 1.3.0
urllib3 1.26.13
wandb 0.22.3
wcwidth 0.2.14
webcolors 25.10.0
webencodings 0.5.1
websocket-client 1.9.0
widgetsnbextension 4.0.15
win32-setctime 1.2.0
xmltodict 1.0.2
yarl 1.22.0

Also, when I check GPU support as recommended in the installation methods, I get the following:
C:\Users\sophi>python -c "import torch; print(torch.cuda.is_available())"
Traceback (most recent call last):
File "", line 1, in
import torch; print(torch.cuda.is_available())
^^^^^^^^^^^^
ModuleNotFoundError: No module named 'torch'

I sense the solution is somewhere near, but I haven't been able to figure it out on my own. Much appreciated again!!
Sophia

[talmo]

Hmmm, so what do you get in the terminal when you do the install? Mine looks like this:

C:\Users\Talmo>uv tool install --python 3.13 "sleap[nn]" --index https://download.pytorch.org/whl/cu130 --index https://pypi.org/simple
Resolved 191 packages in 6.20s
Audited 191 packages in 1ms
Installed 9 executables: sleap-convert, sleap-diagnostic, sleap-inspect, sleap-label, sleap-nn-track, sleap-nn-train, sleap-render, sleap-track, sleap-train

which you can see pulls in the sleap-nn-train CLI.

[sophiaanner]

When I installed it, it said the same as yours. However, I think I have found the solution (with help from claude.ai)! When installing, I specified:
uv tool install --python 3.13 "sleap[nn]==1.5.2" --index https://pypi.org/simple --index https://download.pytorch.org/whl/cu128

Still not 100% sure why this worked vs the actual installation methods listed, but by being explicit about the version and putting the Pypi index first, it installed with no issues and (most importantly) is running training on the centroid model (currently) with no issues!! I also

I will report back if something else comes up, but fingers crossed it's resolved for now and I can move on with my project.

Thanks so much! You and the SLEAP team will be heavily featured in my dissertation acknowledgements

EDIT: spoke too soon

Seems to be only using my CPU, which would explain why it's so slow.

[talmo]

Hi @sophiaanner ,

So just to confirm: you're getting the same error if you use the same exact command I used above? (13.0, pytorch index first)?

Copying it here to be unambiguous:

uv tool install --python 3.13 "sleap[nn]" --index https://download.pytorch.org/whl/cu130 --index https://pypi.org/simple

[sophiaanner]

Hi @talmo,

That's correct. to check, I uninstalled and reinstalled the exact code you just provided. This is what the command line reported after installation:

C:\Users\sophi>uv tool install --python 3.13 "sleap[nn]" --index https://download.pytorch.org/whl/cu130 --index https://pypi.org/simple
Resolved 192 packages in 3.87s
Installed 192 packages in 5.13s

• aiohappyeyeballs==2.6.1
• aiohttp==3.13.2
• aiosignal==1.4.0
• albucore==0.0.24
• albumentations==2.0.8
• annotated-types==0.7.0
• antlr4-python3-runtime==4.9.3
• anyio==4.11.0
• argon2-cffi==25.1.0
• argon2-cffi-bindings==25.1.0
• arrow==1.4.0
• asttokens==3.0.0
• async-lru==2.0.5
• attrs==25.4.0
• av==15.1.0
• babel==2.17.0
• beautifulsoup4==4.14.2
• bleach==6.3.0
• cattrs==25.3.0
• certifi==2022.12.7
• cffi==2.0.0
• charset-normalizer==2.1.1
• click==8.3.0
• colorama==0.4.6
• comm==0.2.3
• contourpy==1.3.3
• cycler==0.12.1
• debugpy==1.8.17
• decorator==5.2.1
• defusedxml==0.7.1
• executing==2.2.1
• fastjsonschema==2.21.2
• filelock==3.19.1
• fonttools==4.60.1
• fqdn==1.5.1
• frozenlist==1.8.0
• fsspec==2025.9.0
• gitdb==4.0.12
• gitpython==3.1.45
• h11==0.16.0
• h5py==3.15.1
• hdmf==4.1.1
• httpcore==1.0.9
• httpx==0.28.1
• hydra-core==1.3.2
• idna==3.4
• imageio==2.37.2
• imageio-ffmpeg==0.6.0
• imgstore==0.3.7
• ipykernel==7.1.0
• ipython==9.6.0
• ipython-pygments-lexers==1.1.1
• ipywidgets==8.1.8
• isoduration==20.11.0
• jedi==0.19.2
• jinja2==3.1.6
• joblib==1.5.2
• jsmin==3.0.1
• json5==0.12.1
• jsonpickle==4.1.1
• jsonpointer==3.0.0
• jsonschema==4.25.1
• jsonschema-specifications==2025.9.1
• jupyter==1.1.1
• jupyter-client==8.6.3
• jupyter-console==6.6.3
• jupyter-core==5.9.1
• jupyter-events==0.12.0
• jupyter-lsp==2.3.0
• jupyter-server==2.17.0
• jupyter-server-terminals==0.5.3
• jupyterlab==4.1.6
• jupyterlab-pygments==0.3.0
• jupyterlab-server==2.24.0
• jupyterlab-widgets==3.0.16
• kiwisolver==1.4.9
• kornia==0.8.1
• kornia-rs==0.1.9
• lark==1.3.1
• lazy-loader==0.4
• lightning==2.5.5
• lightning-utilities==0.11.8
• loguru==0.7.3
• markdown-it-py==4.0.0
• markupsafe==2.1.5
• matplotlib==3.10.7
• matplotlib-inline==0.2.1
• mdurl==0.1.2
• mistune==3.1.4
• mpmath==1.3.0
• multidict==6.7.0
• nbclient==0.10.2
• nbconvert==7.16.6
• nbformat==5.10.4
• ndx-pose==0.2.2
• nest-asyncio==1.6.0
• networkx==3.5
• nixio==1.5.4
• notebook==7.1.3
• notebook-shim==0.2.4
• numpy==2.3.3
• omegaconf==2.3.0
• opencv-python==4.11.0.86
• opencv-python-headless==4.11.0.86
• packaging==24.1
• pandas==2.3.3
• pandocfilters==1.5.1
• parso==0.8.5
• pillow==11.3.0
• platformdirs==4.5.0
• prometheus-client==0.23.1
• prompt-toolkit==3.0.52
• propcache==0.4.1
• protobuf==6.33.0
• psutil==7.1.3
• pure-eval==0.2.3
• pycparser==2.23
• pydantic==2.12.3
• pydantic-core==2.41.4
• pygments==2.19.2
• pykalman==0.10.2
• pymatreader==1.1.0
• pynwb==3.1.2
• pyparsing==3.2.5
• pyside6==6.10.0
• pyside6-addons==6.10.0
• pyside6-essentials==6.10.0
• python-dateutil==2.9.0.post0
• python-json-logger==4.0.0
• python-rapidjson==1.22
• pytorch-lightning==2.5.5
• pytz==2025.2
• pywinpty==3.0.2
• pyyaml==6.0.3
• pyzmq==27.1.0
• qtpy==2.4.3
• referencing==0.37.0
• requests==2.28.1
• rfc3339-validator==0.1.4
• rfc3986-validator==0.1.1
• rfc3987-syntax==1.1.0
• rich==14.2.0
• rich-click==1.9.4
• rpds-py==0.28.0
• ruamel-yaml==0.18.16
• ruamel-yaml-clib==0.2.14
• scikit-base==0.12.6
• scikit-image==0.25.2
• scikit-learn==1.7.2
• scipy==1.16.3
• seaborn==0.13.2
• send2trash==1.8.3
• sentry-sdk==2.43.0
• setuptools==70.2.0
• shiboken6==6.10.0
• simplejson==3.20.2
• simsimd==6.5.3
• six==1.17.0
• sleap==1.5.1
• sleap-io==0.5.7
• sleap-nn==0.0.3
• smmap==5.0.2
• sniffio==1.3.1
• soupsieve==2.8
• stack-data==0.6.3
• stringzilla==4.2.3
• sympy==1.14.0
• terminado==0.18.1
• threadpoolctl==3.6.0
• tifffile==2025.10.16
• tinycss2==1.4.0
• torch==2.9.0+cu130
• torchmetrics==1.0.3
• torchvision==0.2.0
• tornado==6.5.2
• tqdm==4.66.5
• traitlets==5.14.3
• typing-extensions==4.15.0
• typing-inspection==0.4.2
• tzdata==2025.2
• tzlocal==5.3.1
• uri-template==1.3.0
• urllib3==1.26.13
• wandb==0.22.3
• wcwidth==0.2.14
• webcolors==25.10.0
• webencodings==0.5.1
• websocket-client==1.9.0
• widgetsnbextension==4.0.15
• win32-setctime==1.2.0
• xmltodict==1.0.2
• yarl==1.22.0
  Installed 9 executables: sleap-convert, sleap-diagnostic, sleap-inspect, sleap-label, sleap-nn-track, sleap-nn-train, sleap-render, sleap-track, sleap-train

Then, this is the report from sleap-diagnostic. Note that it installed version 1.5.1.
==========SYSTEM==========

utc: 2025-11-05 04:50:58.925553
python: 3.13.8
system: Windows, AMD64, 11, 10.0.26100
path: C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\cv2../../x64/vc14/bin;C:\ffmpeg\bin;C:\windows\system32;C:\windows;C:\windows\System32\Wbem;C:\windows\System32\WindowsPowerShell\v1.0;C:\windows\System32\OpenSSH;C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common;C:\Program Files\dotnet;C:\Users\sophi.local\bin;C:\Users\sophi\AppData\Local\Microsoft\WindowsApps;C:\Users\sophi\AppData\Local\Python\bin;C:\Users\sophi.local\bin;C:\Users\sophi\AppData\Local\Microsoft\WindowsApps;C:\Users\sophi\AppData\Local\Python\bin

==========IMPORTS==========

sleap import: True
sleap path: C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_init_.py
sleap version: 1.5.1
pyside2 import: False
pyside6 import: True
pyside path: C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\PySide6_init_.py
call to C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Scripts\python.exe failed
C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Scripts\python.exe: can't open file 'C:\Users\sophi\.local\bin\sleap-diagnostic': [Errno 2] No such file or directory

cv2 import: True

==========GIT==========

unable to locate git
unable to locate git

==========PYTORCH==========

pytorch import: True
pytorch version: 2.9.0+cu130
pytorch path: C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch_init_.py
cuda available: True
cuda device count: 1
cuda device 0 name: NVIDIA RTX PRO 500 Blackwell Generation Laptop GPU

==========CONDA==========

unable to locate conda

==========PIP==========

unable to locate pip

==========NVIDIA==========

Tue Nov 4 21:51:26 2025
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 581.42 Driver Version: 581.42 CUDA Version: 13.0 |
+-----------------------------------------+------------------------+----------------------+
| GPU Name Driver-Model | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA RTX PRO 500 Black... WDDM | 00000000:01:00.0 Off | N/A |
| N/A 38C P8 3W / 45W | 67MiB / 6113MiB | 1% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| 0 N/A N/A 9108 C+G ...yb3d8bbwe\WindowsTerminal.exe N/A |
+-----------------------------------------------------------------------------------------+

The GUI opens with sleap-label and I am able to open my project. However, when I try to run the training, I get the following error message (as above):
Start training centroid...
['sleap-nn-train', '--config-name', '251104_215348_config', '--config-dir', 'C:\Users\sophi\AppData\Local\Temp\tmpam8rnnjh']
ERROR:sleap.cli:sleap-nn is not installed. This appears to be a GUI-only installation. To enable training, please install SLEAP with the 'nn' dependency. See the installation guide: https://docs.sleap.ai/latest/installation/
Run Path: C:/Users/sophi/OneDrive - University of Louisville/Desktop/models/251104_215348.centroid.n=137
Finished training centroid.
Resetting monitor window.
Start training centered_instance...
['sleap-nn-train', '--config-name', '251104_215452_config', '--config-dir', 'C:\Users\sophi\AppData\Local\Temp\tmp7pulcon5']
ERROR:sleap.cli:sleap-nn is not installed. This appears to be a GUI-only installation. To enable training, please install SLEAP with the 'nn' dependency. See the installation guide: https://docs.sleap.ai/latest/installation/
Run Path: C:/Users/sophi/OneDrive - University of Louisville/Desktop/models/251104_215452.centered_instance.n=137
Finished training centered_instance.
Command line call:
sleap-nn-track --data_path C:/Users/sophi/OneDrive - University of Louisville/Desktop/20251028_MergedProj.slp --video_index 0 --frames 755,4622,5442,5564,7078,7769,8330,9621,10189,14818,19099,21703,25053,26505,26513,28452,30601,37799,38338,41190 --model_paths C:/Users/sophi/OneDrive - University of Louisville/Desktop/models/251104_215348.centroid.n=137 --model_paths C:/Users/sophi/OneDrive - University of Louisville/Desktop/models/251104_215452.centered_instance.n=137 -o C:/Users/sophi/OneDrive - University of Louisville/Desktop\predictions\20251028_MergedProj.slp.251104_215459.predictions.slp

ERROR:sleap.cli:sleap-nn is not installed. This appears to be a GUI-only installation. To enable training, please install SLEAP with the 'nn' dependency. See the installation guide: https://docs.sleap.ai/latest/installation/

Process return code: 0
Traceback (most recent call last):
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\learning\dialog.py", line 763, in run
new_counts = runners.run_learning_pipeline(
labels_filename=self.labels_filename,
...<3 lines>...
items_for_inference=items_for_inference,
)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\learning\runners.py", line 651, in run_learning_pipeline
new_labeled_frame_count = run_gui_inference(inference_task, items_for_inference)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\learning\runners.py", line 877, in run_gui_inference
predictions_path, ret = inference_task.predict_subprocess(
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^
item_for_inference,
^^^^^^^^^^^^^^^^^^^
...<2 lines>...
gui=gui,
^^^^^^^^
)
^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\learning\runners.py", line 401, in predict_subprocess
new_inference_labels = sio.load_slp(output_path)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_io\io\main.py", line 49, in load_slp
return slp.read_labels(filename, open_videos=open_videos)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_io\io\slp.py", line 1877, in read_labels
tracks = read_tracks(labels_path)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_io\io\slp.py", line 808, in read_tracks
tracks = [json.loads(x) for x in read_hdf5_dataset(labels_path, "tracks_json")]
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_io\io\utils.py", line 22, in read_hdf5_dataset
with h5py.File(filename, "r") as f:
~~~~~~~~~^^^^^^^^^^^^^^^
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\h5py_hl\files.py", line 566, in init
fid = make_fid(name, mode, userblock_size, fapl, fcpl, swmr=swmr)
File "C:\Users\sophi\AppData\Roaming\uv\tools\sleap\Lib\site-packages\h5py_hl\files.py", line 241, in make_fid
fid = h5f.open(name, flags, fapl=fapl)
File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper
File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper
File "h5py/h5f.pyx", line 104, in h5py.h5f.open
FileNotFoundError: [Errno 2] Unable to synchronously open file (unable to open file: name = 'C:/Users/sophi/OneDrive - University of Louisville/Desktop\predictions\20251028_MergedProj.slp.251104_215459.predictions.slp', errno = 2, error message = 'No such file or directory', flags = 0, o_flags = 0)

The only fix that allowed me to fully run the training and get predictions was installing with CUDA 12.8 and reordering the indices, as so:
uv tool install --python 3.13 "sleap[nn]==1.5.2" --index https://pypi.org/simple --index https://download.pytorch.org/whl/cu128

But, then when it ran it used the CPU instead of the GPU.

Looking forward to your response,
Sophia

[sophiaanner]

@talmo, any thoughts on solutions? Currently moving forward by running with CPU, but it's incredibly slow as you can imagine.

Thanks!
Sophia