ValueError: zero-size array to reduction operation maximum which has no identity

[sebherbert]

Describe the bug
Running the stardist-napari plugin outputs a value error. Napari doesn's crash and I think the output is still ok though.

(napari-stardist) herber0000@herber0000-desktop:~$ napari
2024-11-12 10:17:35.763860: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2024-11-12 10:17:36.533885: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
2024-11-12 10:17:52.689143: W tensorflow/core/common_runtime/gpu/gpu_device.cc:2251] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
Loading network weights from 'weights_best.h5'.
Couldn't load thresholds from 'thresholds.json', using default values. (Call 'optimize_thresholds' to change that.)
Using default values: prob_thresh=0.5, nms_thresh=0.4.
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
File ~/miniconda3/envs/napari-stardist/lib/python3.10/site-packages/superqt/utils/_qthreading.py:613, in create_worker.<locals>.reraise(e=ValueError('zero-size array to reduction operation maximum which has no identity'))
    612 def reraise(e):
--> 613     raise e
        e = ValueError('zero-size array to reduction operation maximum which has no identity')

File ~/miniconda3/envs/napari-stardist/lib/python3.10/site-packages/superqt/utils/_qthreading.py:175, in WorkerBase.run(self=<napari._qt.qthreading.GeneratorWorker object>)
    173     warnings.filterwarnings("always")
    174     warnings.showwarning = lambda *w: self.warned.emit(w)
--> 175     result = self.work()
        self = <napari._qt.qthreading.GeneratorWorker object at 0x7a6b4d5d7910>
    176 if isinstance(result, Exception):
    177     if isinstance(result, RuntimeError):
    178         # The Worker object has likely been deleted.
    179         # A deleted wrapped C/C++ object may result in a runtime
    180         # error that will cause segfault if we try to do much other
    181         # than simply notify the user.

File ~/miniconda3/envs/napari-stardist/lib/python3.10/site-packages/superqt/utils/_qthreading.py:440, in GeneratorWorker.work(self=<napari._qt.qthreading.GeneratorWorker object>)
    438 try:
    439     _input = self._next_value()
--> 440     output = self._gen.send(_input)
        self = <napari._qt.qthreading.GeneratorWorker object at 0x7a6b4d5d7910>
        _input = None
        output = ('nms', None)
        self._gen = <generator object _plugin_wrapper.<locals>.plugin.<locals>.run at 0x7a6b4ed6e960>
    441     self.yielded.emit(output)
    442 except StopIteration as exc:

File ~/miniconda3/envs/napari-stardist/lib/python3.10/site-packages/stardist_napari/_dock_widget.py:1085, in _plugin_wrapper.<locals>.plugin.<locals>.run()
   1075 @thread_worker(
   1076     connect={
   1077         "yielded": progress_update,
   (...)
   1082 )
   1083 def run():
   1084     r = None
-> 1085     for r in computation_generator:
        r = ('nms', None)
        computation_generator = <generator object _plugin_wrapper.<locals>.plugin_function_generator at 0x7a6b4ed6dee0>
   1086         if is_progress_msg(r):
   1087             yield r

File ~/miniconda3/envs/napari-stardist/lib/python3.10/site-packages/stardist_napari/_dock_widget.py:680, in _plugin_wrapper.<locals>.plugin_function_generator(model=StarDist3D(stardist_23102024_combo108_epoch200_w..., 'patience': 40, 'min_delta': 0}, use_gpu=False), image=<Image layer 'C3-03022022_Label49_t1_100x_0.21_02_POS_current_crop'>, axes='ZYX', slice_image=None, norm_image=True, perc_low=1.0, perc_high=99.8, input_scale=None, prob_thresh=0.5, nms_thresh=0.4, output_type=<Output.Polys: 'Polygons / Polyhedra'>, n_tiles=None, norm_axes='ZYX', timelapse_opts=<TimelapseLabels.Unique: 'Unique through time'>, cnn_output=False, image_data=array([[[56, 26, ..., 39, 33],
        [47, 56, ...],
        [40, 52, ..., 65, 35]]], dtype=uint16))
    672     n_objects = len(polys["points"])
    673     surface = surface_from_polys(polys)
    674     layers.append(
    675         move_layer_axes(
    676             (
    677                 surface,
    678                 dict(
    679                     name="StarDist polyhedra",
--> 680                     contrast_limits=(0, surface[-1].max()),
        layers = []
        surface = [array([], dtype=float64), array([], dtype=float64), array([], dtype=float64)]
        surface[-1] = array([], dtype=float64)
        scale_out = [1.0, 1.0, 1.0]
        n_objects = 0
        origin_out = [0, 0, 0]
        layer_axes_from = 'ZYX'
        layer_axes_to = 'ZYX'
    681                     scale=scale_out,
    682                     colormap=label_colormap(n_objects),
    683                     translate=origin_out,
    684                 ),
    685                 "surface",
    686             ),
    687             layer_axes_from,
    688             layer_axes_to,
    689         )
    690     )
    691 else:
    692     # TODO: coordinates correct or need offset (0.5 or so)?
    693     shapes = np.moveaxis(polys["coord"], -1, -2)

File ~/miniconda3/envs/napari-stardist/lib/python3.10/site-packages/numpy/core/_methods.py:41, in _amax(a=array([], dtype=float64), axis=None, out=None, keepdims=False, initial=<no value>, where=True)
     39 def _amax(a, axis=None, out=None, keepdims=False,
     40           initial=_NoValue, where=True):
---> 41     return umr_maximum(a, axis, None, out, keepdims, initial, where)
        umr_maximum = <built-in method reduce of numpy.ufunc object at 0x7a6bf747ce40>
        a = array([], dtype=float64)
        axis = None
        out = None
        keepdims = False
        initial = <no value>
        where = True

ValueError: zero-size array to reduction operation maximum which has no identity

To reproduce
Python env
Create a conda environment with python 3.10
pip install tensorflow[and-cuda] (GPU compatibility test )
pip install stardist
pip install stardist-napari
pip install PyQt5
pip install tensorflow==2.16.1 (to fix the numpy crash in v2.1.0, downgrade to 1.24 is only possible if downgrade for tf 2.18.0 to 2.16.1)
stardist-napari now boots in napari

Expected behavior
No erro message

Data and screenshots
Test dataset and custom network can be found here: https://drive.switch.ch/index.php/s/gxpHKj0Pg4esmE2

Environment:

1. Please copy and paste the information at napari info option in the help menubar.

napari: 0.5.4
Platform: Linux-6.8.0-48-generic-x86_64-with-glibc2.39
System: Ubuntu 24.04.1 LTS
Python: 3.10.15 | packaged by conda-forge | (main, Oct 16 2024, 01:24:24) [GCC 13.3.0]
Qt: 5.15.14
PyQt5: 5.15.11
NumPy: 1.24.0
SciPy: 1.14.1
Dask: 2024.11.0
VisPy: 0.14.3
magicgui: 0.9.1
superqt: 0.6.7
in-n-out: 0.2.1
app-model: 0.3.0
psygnal: 0.11.1
npe2: 0.7.7
pydantic: 2.9.2

OpenGL:
- GL version: 4.6.0 NVIDIA 535.183.01
- MAX_TEXTURE_SIZE: 32768
- GL_MAX_3D_TEXTURE_SIZE: 16384

Screens:
- screen 1: resolution 1920x1200, scale 1.0
- screen 2: resolution 1600x1200, scale 1.0

Optional:
- numba: 0.60.0
- triangle not installed
- napari-plugin-manager not installed

Settings path:
- /home/herber0000/.config/napari/napari-stardist_3a606daac139f953f571ad550336c6fe9e3198c3/settings.yaml

2. Please run this code and paste the output:

os: Linux-6.8.0-48-generic-x86_64-with-glibc2.39
stardist_napari: 2024.8.6.1
stardist: 0.9.1
csbdeep: 0.8.1
napari: 0.5.4
magicgui: 0.9.1
2024-11-12 14:22:47.855205: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2024-11-12 14:22:48.477994: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
tensorflow: 2.16.1