Inference step fails - Githubissues

dzenanz commented 10 months ago

I followed the installation and usage instructions, only to get the below errors in the log when clicking "Segmentation" button. This happens with both Slicer's sample data "CBCT-MR Head, and with my private data.

sending infer request...
Starting new HTTP connection (1): 127.0.0.1:5555
93 [0, 134, 86, 239] [58, 128]
INFO:     127.0.0.1:56739 - "POST /infer HTTP/1.1" 500 Internal Server Error
http://127.0.0.1:5555 "POST /infer HTTP/1.1" 500 21
ERROR:    Exception in ASGI application
Traceback (most recent call last):
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\uvicorn\protocols\http\h11_impl.py", line 404, in run_asgi
    result = await app(  # type: ignore[func-returns-value]
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\uvicorn\middleware\proxy_headers.py", line 84, in __call__
Traceback (most recent call last):
    return await self.app(scope, receive, send)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\fastapi\applications.py", line 1054, in __call__
  File "C:/Dev/Work/MedSAMSlicer-0.01/MedSAM/MedSAMLite/MedSAMLite.py", line 236, in <lambda>
    await super().__call__(scope, receive, send)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\applications.py", line 123, in __call__
    await self.middleware_stack(scope, receive, send)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\middleware\errors.py", line 186, in __call__
    raise exc
    self.ui.pbSegment.connect('clicked(bool)', lambda: self.logic.applySegmentation())
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\middleware\errors.py", line 164, in __call__
  File "C:/Dev/Work/MedSAMSlicer-0.01/MedSAM/MedSAMLite/MedSAMLite.py", line 606, in applySegmentation
    await self.app(scope, receive, _send)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\middleware\exceptions.py", line 62, in __call__
    await wrap_app_handling_exceptions(self.app, conn)(scope, receive, send)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\_exception_handler.py", line 64, in wrapped_app
    raise exc
    segmentation_mask = self.inferSegmentation(serverUrl)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\_exception_handler.py", line 53, in wrapped_app
  File "C:/Dev/Work/MedSAMSlicer-0.01/MedSAM/MedSAMLite/MedSAMLite.py", line 579, in inferSegmentation
    await app(scope, receive, sender)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\routing.py", line 762, in __call__
    await self.middleware_stack(scope, receive, send)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\routing.py", line 782, in app
    await route.handle(scope, receive, send)
    response.raise_for_status()
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\routing.py", line 297, in handle
    await self.app(scope, receive, send)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\requests\models.py", line 1021, in raise_for_status
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\routing.py", line 77, in app
    await wrap_app_handling_exceptions(app, request)(scope, receive, send)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\_exception_handler.py", line 64, in wrapped_app
    raise exc
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\_exception_handler.py", line 53, in wrapped_app
    await app(scope, receive, sender)
    raise HTTPError(http_error_msg, response=self)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\routing.py", line 72, in app
    response = await func(request)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\fastapi\routing.py", line 299, in app
    raise e
requests.exceptions.HTTPError: 500 Server Error: Internal Server Error for url: http://127.0.0.1:5555/infer
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\fastapi\routing.py", line 294, in app
    raw_response = await run_endpoint_function(
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\fastapi\routing.py", line 193, in run_endpoint_function
    return await run_in_threadpool(dependant.call, **values)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\starlette\concurrency.py", line 40, in run_in_threadpool
    return await anyio.to_thread.run_sync(func, *args)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\anyio\to_thread.py", line 56, in run_sync
    return await get_async_backend().run_sync_in_worker_thread(
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\anyio\_backends\_asyncio.py", line 2134, in run_sync_in_worker_thread
    return await future
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\anyio\_backends\_asyncio.py", line 851, in run
    result = context.run(func, *args)
  File "C:\Dev\Work\MedSAMSlicer-0.01\_deps\server_essentials\server.py", line 316, in infer
    res[csidx], bbox_1024_prev = inf(bbox_1024_prev)
  File "C:\Dev\Work\MedSAMSlicer-0.01\_deps\server_essentials\server.py", line 313, in inf
    return mask.tolist(), [get_bbox1024(mask_t)]
  File "C:\Dev\Work\MedSAMSlicer-0.01\_deps\server_essentials\server.py", line 263, in get_bbox1024
    x_min, x_max = np.min(x_indices), np.max(x_indices)
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\numpy\core\fromnumeric.py", line 2953, in min
    return _wrapreduction(a, np.minimum, 'min', axis, None, out,
  File "C:\Users\Dzenan\AppData\Local\slicer.org\Slicer 5.6.1\lib\Python\Lib\site-packages\numpy\core\fromnumeric.py", line 88, in _wrapreduction
    return ufunc.reduce(obj, axis, dtype, out, **passkwargs)
ValueError: zero-size array to reduction operation minimum which has no identity

rasakereh commented 10 months ago

Thank you for sharing the issue with us. Would you please follow these steps and send me a screenshot:

Five buttons to the right of the drop-down menu click on the cube-like icon with tooltip text "Volumes".
For Window/Level value select Manual Min/Max and the range of 0 to 255.

It will show how our plugin actually sees your image. I guess the error happens because there is not much separation between the segment you want to identify and the background. As a temporary fix you can try applying either of the preprocessing steps to increase the color contrasts. We will work on it to fix the issue and/or introduce more relevant preprocessing options.

Let me know if it fixed the problem please.

Thanks!

tejoman commented 10 months ago

For what its worth I had this same issue and it resolved after preprocessing my data to 0 to 255 intensity range as described by rasakereh.

neel-dey commented 10 months ago

On Linux and facing the same issue. Renormalizing to [0, 255] did not resolve this issue, see for example:

File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/starlette/routing.py", line 782, in app
    await route.handle(scope, receive, send)
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/starlette/routing.py", line 297, in handle
    await self.app(scope, receive, send)
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/starlette/routing.py", line 77, in app
    await wrap_app_handling_exceptions(app, request)(scope, receive, send)
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/starlette/_exception_handler.py", line 64, in wrapped_app
    raise exc
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/starlette/_exception_handler.py", line 53, in wrapped_app
    await app(scope, receive, sender)
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/starlette/routing.py", line 72, in app
    response = await func(request)
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/fastapi/routing.py", line 299, in app
    raise e
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/fastapi/routing.py", line 294, in app
    raw_response = await run_endpoint_function(
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/fastapi/routing.py", line 193, in run_endpoint_function
    return await run_in_threadpool(dependant.call, **values)
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/starlette/concurrency.py", line 40, in run_in_threadpool
    return await anyio.to_thread.run_sync(func, *args)
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/anyio/to_thread.py", line 56, in run_sync
    return await get_async_backend().run_sync_in_worker_thread(
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/anyio/_backends/_asyncio.py", line 2134, in run_sync_in_worker_thread
    return await future
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/anyio/_backends/_asyncio.py", line 851, in run
    result = context.run(func, *args)
  File "/home/neel/Desktop/MedSAMSlicer-0.01/dependencies/server_essentials/server.py", line 316, in infer
    res[csidx], bbox_1024_prev = inf(bbox_1024_prev)
  File "/home/neel/Desktop/MedSAMSlicer-0.01/dependencies/server_essentials/server.py", line 313, in inf
    return mask.tolist(), [get_bbox1024(mask_t)]
  File "/home/neel/Desktop/MedSAMSlicer-0.01/dependencies/server_essentials/server.py", line 263, in get_bbox1024
    x_min, x_max = np.min(x_indices), np.max(x_indices)
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/numpy/core/fromnumeric.py", line 2953, in min
    return _wrapreduction(a, np.minimum, 'min', axis, None, out,
  File "/home/neel/Desktop/Slicer-5.6.1-linux-amd64/lib/Python/lib/python3.9/site-packages/numpy/core/fromnumeric.py", line 88, in _wrapreduction
    return ufunc.reduce(obj, axis, dtype, out, **passkwargs)
ValueError: zero-size array to reduction operation minimum which has no identity

rasakereh commented 9 months ago

A new release is ready. We tried to address this issue. Please remove pre-existing version and install the new one and let me know if it fixed your issue.

neel-dey commented 9 months ago

Thanks for the new release. There's now a different error when using the same image.

Traceback (most recent call last):
  File "/home/neel/Desktop/MedSAMSlicer-0.02/MedSAM/MedSAMLite/MedSAMLite.py", line 237, in <lambda>
    self.ui.pbSegment.connect('clicked(bool)', lambda: self.logic.applySegmentation())
  File "/home/neel/Desktop/MedSAMSlicer-0.02/MedSAM/MedSAMLite/MedSAMLite.py", line 634, in applySegmentation
    segmentation_mask = self.inferSegmentation(serverUrl)
  File "/home/neel/Desktop/MedSAMSlicer-0.02/MedSAM/MedSAMLite/MedSAMLite.py", line 608, in inferSegmentation
    seg_result[frame, :, :] = seg_data[str(frame)]
ValueError: could not broadcast input array from shape (150,146) into shape (146,150)

Looking at the trace, I'm guessing a transpose is missing somewhere?

Thanks for the help!

rasakereh commented 9 months ago

@neel-dey if there is no privacy concern, can you please share the image file with me? my email: reza.asakereh@uhn.ca

neel-dey commented 9 months ago

@rasakereh Oh sure, it's public data available from here

FYI, I just tried with a different (160, 160, 192) brain volume and was able to get an output whereas the image I linked to is (150, 146, 249) volume. Perhaps there's a divisibility by powers of two or resizing error happening somewhere?

rasakereh commented 9 months ago

Thanks @neel-dey . You are right, there was a faulty dimension swap in the server. It is now fixed in the new release. Also we implemented a better upgrade solution in the module so you dont have to do it manually each time. Hope it fixes the bug.

neel-dey commented 9 months ago

The interface does produce outputs now, thanks! I didn't create this issue so I can't close it, but this particular issue seems resolved from my side.

dzenanz commented 9 months ago

I am running into a different issue now:

sending setImage request...
Starting new HTTP connection (1): 127.0.0.1:5555
http://127.0.0.1:5555 "POST /setImage HTTP/1.1" 200 4
Response from setImage: null
Traceback (most recent call last):
  File "C:/Dev/Work/MedSAMSlicer/MedSAM/MedSAMLite/MedSAMLite.py", line 243, in <lambda>
    self.ui.pbSendImage.connect('clicked(bool)', lambda: self.logic.sendImage())
  File "C:/Dev/Work/MedSAMSlicer/MedSAM/MedSAMLite/MedSAMLite.py", line 642, in sendImage
    self.captureImage()
  File "C:/Dev/Work/MedSAMSlicer/MedSAM/MedSAMLite/MedSAMLite.py", line 631, in captureImage
    self.img_path = self.volume_node.GetStorageNode().GetFullNameFromFileName()
AttributeError: 'NoneType' object has no attribute 'GetFullNameFromFileName'

rasakereh commented 9 months ago

@dzenanz refer to issue #6 for this. You have to open an actual file (sth that exists on the disk) for your segmentation task

dzenanz commented 9 months ago

Close this issue, if you think it is fixed.

bowang-lab / MedSAMSlicer

Inference step fails #1