nod-ai / SHARK

SHARK - High Performance Machine Learning Distribution
Apache License 2.0
1.4k stars 169 forks source link

[BUG][Pre-release] ImportError: cannot import name 'ParameterIndexEntry' from 'iree.runtime' #2117

Open unsivilaudio opened 2 months ago

unsivilaudio commented 2 months ago

OS: Windows 11 GPU: AMD Radeon 7900xtx DRIVER: 24.1.1

Fails to start with stacktrace

vulkan devices are available.
metal devices are not available.
cuda devices are not available.
rocm devices are not available.
local-sync devices are available.
local-task devices are available.
Clearing .mlir temporary files from a prior run. This may take some time...
Clearing .mlir temporary files took 0.0000 seconds.
gradio temporary image cache located at X:\shark-ai\shark_tmp/gradio. You may change this by setting the GRADIO_TEMP_DIR environment variable.
No temporary images files to clear.
gradio temporary image cache located at X:\shark-ai\shark_tmp/gradio. You may change this by setting the GRADIO_TEMP_DIR environment variable.
No temporary images files to clear.
Traceback (most recent call last):
  File "index.py", line 222, in <module>
  File "index.py", line 86, in webui
  File "PyInstaller\loader\pyimod02_importers.py", line 419, in exec_module
  File "ui\chat.py", line 8, in <module>
  File "PyInstaller\loader\pyimod02_importers.py", line 419, in exec_module
  File "apps\shark_studio\api\llm.py", line 1, in <module>
  File "PyInstaller\loader\pyimod02_importers.py", line 419, in exec_module
  File "turbine_models\custom_models\stateless_llama.py", line 11, in <module>
  File "PyInstaller\loader\pyimod02_importers.py", line 419, in exec_module
  File "shark_turbine\aot\__init__.py", line 13, in <module>
  File "PyInstaller\loader\pyimod02_importers.py", line 419, in exec_module
  File "shark_turbine\aot\params.py", line 15, in <module>
ImportError: cannot import name 'ParameterIndexEntry' from 'iree.runtime' (C:\Users\Trinity\AppData\Local\Temp\_MEI7084882\iree\runtime\__init__.pyc)

Not sure if you were aware, just putting this out there to get checked. Will go back to release version in the meantime. :)