RuntimeError: max(): Expected reduction dim to be specified for input.numel() == 0. Specify the reduction dim with the 'dim' argument.

It seems that there is something wrong with coords

D:\TRELLIS_env\Scripts\python.exe D:\TRELLIS\app_text.py 
[SPARSE] Backend: spconv, Attention: flash_attn
Warp 1.5.0 initialized:
   CUDA Toolkit 12.6, Driver 12.4
   Devices:
     "cpu"      : "Intel64 Family 6 Model 143 Stepping 8, GenuineIntel"
     "cuda:0"   : "NVIDIA RTX 5880 Ada Generation" (48 GiB, sm_89, mempool enabled)
   Kernel cache:
     C:\Users\Admin\AppData\Local\NVIDIA\warp\Cache\1.5.0
D:\TRELLIS_env\Lib\site-packages\gradio_client\utils.py:1097: UserWarning: file() is deprecated and will be removed in a future version. Use handle_file() instead.
  warnings.warn(
[SPARSE][CONV] spconv algo: auto
[ATTENTION] Using backend: flash_attn
D:\TRELLIS_env\Lib\site-packages\transformers\models\clip\modeling_clip.py:540: UserWarning: 1Torch was not compiled with flash attention. (Triggered internally at C:\actions-runner\_work\pytorch\pytorch\builder\windows\pytorch\aten\src\ATen\native\transformers\cuda\sdp_utils.cpp:555.)
  attn_output = torch.nn.functional.scaled_dot_product_attention(
Running on local URL:  http://0.0.0.0:9101

Could not create share link. Please check your internet connection or our status page: https://status.gradio.app.
Sampling: 100%|██████████| 25/25 [00:03<00:00,  7.67it/s]
Traceback (most recent call last):
  File "D:\TRELLIS_env\Lib\site-packages\gradio\queueing.py", line 536, in process_events
    response = await route_utils.call_process_api(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\TRELLIS_env\Lib\site-packages\gradio\route_utils.py", line 322, in call_process_api
    output = await app.get_blocks().process_api(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\TRELLIS_env\Lib\site-packages\gradio\blocks.py", line 1935, in process_api
    result = await self.call_function(
             ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\TRELLIS_env\Lib\site-packages\gradio\blocks.py", line 1520, in call_function
    prediction = await anyio.to_thread.run_sync(  # type: ignore
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\TRELLIS_env\Lib\site-packages\anyio\to_thread.py", line 56, in run_sync
    return await get_async_backend().run_sync_in_worker_thread(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\TRELLIS_env\Lib\site-packages\anyio\_backends\_asyncio.py", line 2505, in run_sync_in_worker_thread
    return await future
           ^^^^^^^^^^^^
  File "D:\TRELLIS_env\Lib\site-packages\anyio\_backends\_asyncio.py", line 1005, in run
    result = context.run(func, *args)
             ^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\TRELLIS_env\Lib\site-packages\gradio\utils.py", line 826, in wrapper
    response = f(*args, **kwargs)
               ^^^^^^^^^^^^^^^^^^
  File "D:\TRELLIS\app_text.py", line 103, in text_to_3d
    outputs = pipeline.run(
              ^^^^^^^^^^^^^
  File "D:\TRELLIS_env\Lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "D:\TRELLIS\trellis\pipelines\trellis_text_to_3d.py", line 227, in run
    slat = self.sample_slat(cond, coords, slat_sampler_params)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\TRELLIS\trellis\pipelines\trellis_text_to_3d.py", line 184, in sample_slat
    noise = sp.SparseTensor(
            ^^^^^^^^^^^^^^^^
  File "D:\TRELLIS\trellis\modules\sparse\basic.py", line 70, in __init__
    shape = self.__cal_shape(feats, coords)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\TRELLIS\trellis\modules\sparse\basic.py", line 119, in __cal_shape
    shape.append(coords[:, 0].max().item() + 1)
                 ^^^^^^^^^^^^^^^^^^
RuntimeError: max(): Expected reduction dim to be specified for input.numel() == 0. Specify the reduction dim with the 'dim' argument.

Start with app_text.py at text_to_3d, then follow trellis/pipelines/trellis_text_to_3d.py through run and sample_slat into trellis/modules/sparse/basic.py::__cal_shape. Reproduce the reported text-to-3d run and inspect why coords is empty before coords[:, 0].max(). Done means the reported RuntimeError no longer occurs and the pipeline completes successfully for the reproduction.