Skip to content

When passing an IntervalView to CSBDeep, the lower limits of the interval seem not to be taken into account correctly. #10

New issue
New issue
Open
Open
When passing an IntervalView to CSBDeep, the lower limits of the interval seem not to be taken into account correctly.#10
@michaelmell

Description

@michaelmell
michaelmell
opened on Jul 10, 2019

When passing an IntervalView to CSBDeep, the lower limits of the interval seem not to be taken into account correctly.

This works for a model with input-tensor size 32 x 512:

FinalInterval roiForNetworkProcessing = new FinalInterval(new long[]{37, img.dimension(1) - 512, 0}, new long[]{68, img.dimension(1) - 1, img.dimension(2) - 1});
IntervalView<FloatType> newImg = Views.interval(img, roiForNetworkProcessing);

Dataset dataset = datasetService.create(Views.zeroMin(newImg)); // WHY DO WE NEED ZEROMIN HERE?!

final CommandModule module = commandService.run(
    GenericNetwork.class, false,
    "input", dataset,
    "modelFile", "/path/to//model.zip",
    "normalizeInput", false,
    "blockMultiple", 8,
    "nTiles", 1,
    "showProgressDialog", true).get();
Img<FloatType> tmp = (Img<FloatType>) module.getOutput("output");

But changing the line

Dataset dataset = datasetService.create(Views.zeroMin(newImg)); // WHY DO WE NEED ZEROMIN HERE?!

into

Dataset dataset = datasetService.create(newImg); // WHY DO WE NEED ZEROMIN HERE?!

causes the error/exception below, where the reported network input size is noteworthy:

[INFO] Network input size: [69, 531, 1, 1]

From this, it seems that only the upper limits of the IntervalView are taken into account.
Note that the size of the input image is 113 x 531.

This complete exception output:

[INFO] Shape of input tensor: [-1, 512, 32, 1]
[INFO] Shape of output tensor: [-1, 512, 32, 1]
[INFO] Dataset type: 32-bit signed float, converting to FloatType.
[INFO] Dataset dimensions: [32, 512, 6]
[INFO] INPUT NODE:
[INFO] Mapping of tensor input:
[INFO]    datasetAxes:[X, Y, Z]
[INFO]    nodeAxes:[(Z, -1), (Y, 512), (X, 32), (Channel, 1)]
[INFO]    mapping:[2, 1, 0, 3]
[INFO] OUTPUT NODE:
[INFO] Mapping of tensor output:
[INFO]    datasetAxes:[X, Y, Z]
[INFO]    nodeAxes:[(Z, -1), (Y, 512), (X, 32), (Channel, 1)]
[INFO]    mapping:[2, 1, 0, 3]
[INFO] Complete input axes: [X, Y, Z, Channel]
[INFO] Tiling actions: [NO_TILING, NO_TILING, TILE_WITHOUT_PADDING, NO_TILING]
[INFO] Dividing image into 6 tile(s)..
[INFO] Size of single image tile: [32, 512, 1, 1]
[INFO] Final image tiling: [1, 1, 6, 1]
[INFO] Network input size: [69, 531, 1, 1]
[INFO] Processing tile 1..
java.util.concurrent.ExecutionException: java.lang.RuntimeException: java.lang.RuntimeException: java.util.concurrent.ExecutionException: java.lang.IllegalArgumentException: ConcatOp : Dimensions of inputs should match: shape[0] = [1,132,16,128] vs. shape[1] = [1,132,17,128]
	 [[{{node concatenate_2/concat}} = ConcatV2[N=2, T=DT_FLOAT, Tidx=DT_INT32, _output_shapes=[[?,128,8,256]], _device="/job:localhost/replica:0/task:0/device:CPU:0"](conv2d_transpose_2/BiasAdd, conv2d_6/Relu, concatenate_1/concat/axis)]]
	at java.util.concurrent.ForkJoinTask.get(ForkJoinTask.java:1006)
	at de.csbdresden.csbdeep.network.DefaultModelExecutor.run(DefaultModelExecutor.java:82)
	at de.csbdresden.csbdeep.network.DefaultModelExecutor.run(DefaultModelExecutor.java:43)
	at de.csbdresden.csbdeep.commands.GenericNetwork.tileAndRunNetwork(GenericNetwork.java:559)
	at de.csbdresden.csbdeep.commands.GenericNetwork.tryToTileAndRunNetwork(GenericNetwork.java:537)
	at de.csbdresden.csbdeep.commands.GenericNetwork.mainThread(GenericNetwork.java:449)
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
	at java.util.concurrent.FutureTask.run$$$capture(FutureTask.java:266)
	at java.util.concurrent.FutureTask.run(FutureTask.java)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:748)
Caused by: java.lang.RuntimeException: java.lang.RuntimeException: java.util.concurrent.ExecutionException: java.lang.IllegalArgumentException: ConcatOp : Dimensions of inputs should match: shape[0] = [1,132,16,128] vs. shape[1] = [1,132,17,128]
	 [[{{node concatenate_2/concat}} = ConcatV2[N=2, T=DT_FLOAT, Tidx=DT_INT32, _output_shapes=[[?,128,8,256]], _device="/job:localhost/replica:0/task:0/device:CPU:0"](conv2d_transpose_2/BiasAdd, conv2d_6/Relu, concatenate_1/concat/axis)]]
	at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
	at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:62)
	at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
	at java.lang.reflect.Constructor.newInstance(Constructor.java:423)
	at java.util.concurrent.ForkJoinTask.getThrowableException(ForkJoinTask.java:593)
	at java.util.concurrent.ForkJoinTask.get(ForkJoinTask.java:1005)
	... 11 more
Caused by: java.lang.RuntimeException: java.util.concurrent.ExecutionException: java.lang.IllegalArgumentException: ConcatOp : Dimensions of inputs should match: shape[0] = [1,132,16,128] vs. shape[1] = [1,132,17,128]
	 [[{{node concatenate_2/concat}} = ConcatV2[N=2, T=DT_FLOAT, Tidx=DT_INT32, _output_shapes=[[?,128,8,256]], _device="/job:localhost/replica:0/task:0/device:CPU:0"](conv2d_transpose_2/BiasAdd, conv2d_6/Relu, concatenate_1/concat/axis)]]
	at java.util.concurrent.ForkJoinTask$AdaptedCallable.exec(ForkJoinTask.java:1431)
	at java.util.concurrent.ForkJoinTask.doExec(ForkJoinTask.java:289)
	at java.util.concurrent.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:1056)
	at java.util.concurrent.ForkJoinPool.runWorker(ForkJoinPool.java:1692)
	at java.util.concurrent.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:157)
Caused by: java.util.concurrent.ExecutionException: java.lang.IllegalArgumentException: ConcatOp : Dimensions of inputs should match: shape[0] = [1,132,16,128] vs. shape[1] = [1,132,17,128]
	 [[{{node concatenate_2/concat}} = ConcatV2[N=2, T=DT_FLOAT, Tidx=DT_INT32, _output_shapes=[[?,128,8,256]], _device="/job:localhost/replica:0/task:0/device:CPU:0"](conv2d_transpose_2/BiasAdd, conv2d_6/Relu, concatenate_1/concat/axis)]]
	at java.util.concurrent.FutureTask.report(FutureTask.java:122)
	at java.util.concurrent.FutureTask.get(FutureTask.java:192)
	at de.csbdresden.csbdeep.network.model.DefaultNetwork.call(DefaultNetwork.java:86)
	at de.csbdresden.csbdeep.network.model.DefaultNetwork.call(DefaultNetwork.java:23)
	at java.util.concurrent.ForkJoinTask$AdaptedCallable.exec(ForkJoinTask.java:1424)
	... 4 more
Caused by: java.lang.IllegalArgumentException: ConcatOp : Dimensions of inputs should match: shape[0] = [1,132,16,128] vs. shape[1] = [1,132,17,128]
	 [[{{node concatenate_2/concat}} = ConcatV2[N=2, T=DT_FLOAT, Tidx=DT_INT32, _output_shapes=[[?,128,8,256]], _device="/job:localhost/replica:0/task:0/device:CPU:0"](conv2d_transpose_2/BiasAdd, conv2d_6/Relu, concatenate_1/concat/axis)]]
	at org.tensorflow.Session.run(Native Method)
	at org.tensorflow.Session.access$100(Session.java:48)
	at org.tensorflow.Session$Runner.runHelper(Session.java:314)
	at org.tensorflow.Session$Runner.run(Session.java:264)
	at de.csbdresden.csbdeep.network.model.tensorflow.TensorFlowRunner.executeGraph(TensorFlowRunner.java:54)
	at de.csbdresden.csbdeep.network.model.tensorflow.TensorFlowNetwork.execute(TensorFlowNetwork.java:345)
	at de.csbdresden.csbdeep.network.model.DefaultNetwork.lambda$call$0(DefaultNetwork.java:78)
	at java.util.concurrent.FutureTask.run$$$capture(FutureTask.java:266)
	at java.util.concurrent.FutureTask.run(FutureTask.java)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:748)
java.util.concurrent.ExecutionException: java.lang.RuntimeException: java.lang.RuntimeException: java.util.concurrent.ExecutionException: java.lang.IllegalArgumentException: ConcatOp : Dimensions of inputs should match: shape[0] = [1,132,16,128] vs. shape[1] = [1,132,17,128]
	 [[{{node concatenate_2/concat}} = ConcatV2[N=2, T=DT_FLOAT, Tidx=DT_INT32, _output_shapes=[[?,128,8,256]], _device="/job:localhost/replica:0/task:0/device:CPU:0"](conv2d_transpose_2/BiasAdd, conv2d_6/Relu, concatenate_1/concat/axis)]]
	at java.util.concurrent.ForkJoinTask.get(ForkJoinTask.java:1006)
	at de.csbdresden.csbdeep.network.DefaultModelExecutor.run(DefaultModelExecutor.java:82)
	at de.csbdresden.csbdeep.network.DefaultModelExecutor.run(DefaultModelExecutor.java:43)
	at de.csbdresden.csbdeep.commands.GenericNetwork.tileAndRunNetwork(GenericNetwork.java:559)
	at de.csbdresden.csbdeep.commands.GenericNetwork.tryToTileAndRunNetwork(GenericNetwork.java:537)
	at de.csbdresden.csbdeep.commands.GenericNetwork.mainThread(GenericNetwork.java:449)
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
	at java.util.concurrent.FutureTask.run$$$capture(FutureTask.java:266)
	at java.util.concurrent.FutureTask.run(FutureTask.java)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:748)
Caused by: java.lang.RuntimeException: java.lang.RuntimeException: java.util.concurrent.ExecutionException: java.lang.IllegalArgumentException: ConcatOp : Dimensions of inputs should match: shape[0] = [1,132,16,128] vs. shape[1] = [1,132,17,128]
	 [[{{node concatenate_2/concat}} = ConcatV2[N=2, T=DT_FLOAT, Tidx=DT_INT32, _output_shapes=[[?,128,8,256]], _device="/job:localhost/replica:0/task:0/device:CPU:0"](conv2d_transpose_2/BiasAdd, conv2d_6/Relu, concatenate_1/concat/axis)]]
	at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
	at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:62)
	at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
	at java.lang.reflect.Constructor.newInstance(Constructor.java:423)
	at java.util.concurrent.ForkJoinTask.getThrowableException(ForkJoinTask.java:593)
	at java.util.concurrent.ForkJoinTask.get(ForkJoinTask.java:1005)
	... 11 more
Caused by: java.lang.RuntimeException: java.util.concurrent.ExecutionException: java.lang.IllegalArgumentException: ConcatOp : Dimensions of inputs should match: shape[0] = [1,132,16,128] vs. shape[1] = [1,132,17,128]
	 [[{{node concatenate_2/concat}} = ConcatV2[N=2, T=DT_FLOAT, Tidx=DT_INT32, _output_shapes=[[?,128,8,256]], _device="/job:localhost/replica:0/task:0/device:CPU:0"](conv2d_transpose_2/BiasAdd, conv2d_6/Relu, concatenate_1/concat/axis)]]
	at java.util.concurrent.ForkJoinTask$AdaptedCallable.exec(ForkJoinTask.java:1431)
	at java.util.concurrent.ForkJoinTask.doExec(ForkJoinTask.java:289)
	at java.util.concurrent.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:1056)
	at java.util.concurrent.ForkJoinPool.runWorker(ForkJoinPool.java:1692)
	at java.util.concurrent.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:157)
Caused by: java.util.concurrent.ExecutionException: java.lang.IllegalArgumentException: ConcatOp : Dimensions of inputs should match: shape[0] = [1,132,16,128] vs. shape[1] = [1,132,17,128]
	 [[{{node concatenate_2/concat}} = ConcatV2[N=2, T=DT_FLOAT, Tidx=DT_INT32, _output_shapes=[[?,128,8,256]], _device="/job:localhost/replica:0/task:0/device:CPU:0"](conv2d_transpose_2/BiasAdd, conv2d_6/Relu, concatenate_1/concat/axis)]]
	at java.util.concurrent.FutureTask.report(FutureTask.java:122)
	at java.util.concurrent.FutureTask.get(FutureTask.java:192)
	at de.csbdresden.csbdeep.network.model.DefaultNetwork.call(DefaultNetwork.java:86)
	at de.csbdresden.csbdeep.network.model.DefaultNetwork.call(DefaultNetwork.java:23)
	at java.util.concurrent.ForkJoinTask$AdaptedCallable.exec(ForkJoinTask.java:1424)
	... 4 more
Caused by: java.lang.IllegalArgumentException: ConcatOp : Dimensions of inputs should match: shape[0] = [1,132,16,128] vs. shape[1] = [1,132,17,128]
	 [[{{node concatenate_2/concat}} = ConcatV2[N=2, T=DT_FLOAT, Tidx=DT_INT32, _output_shapes=[[?,128,8,256]], _device="/job:localhost/replica:0/task:0/device:CPU:0"](conv2d_transpose_2/BiasAdd, conv2d_6/Relu, concatenate_1/concat/axis)]]
	at org.tensorflow.Session.run(Native Method)
	at org.tensorflow.Session.access$100(Session.java:48)
	at org.tensorflow.Session$Runner.runHelper(Session.java:314)
	at org.tensorflow.Session$Runner.run(Session.java:264)
	at de.csbdresden.csbdeep.network.model.tensorflow.TensorFlowRunner.executeGraph(TensorFlowRunner.java:54)
	at de.csbdresden.csbdeep.network.model.tensorflow.TensorFlowNetwork.execute(TensorFlowNetwork.java:345)
	at de.csbdresden.csbdeep.network.model.DefaultNetwork.lambda$call$0(DefaultNetwork.java:78)
	at java.util.concurrent.FutureTask.run$$$capture(FutureTask.java:266)
	at java.util.concurrent.FutureTask.run(FutureTask.java)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:748)
[INFO] Plugin exit (took 1273 milliseconds)
java.lang.NullPointerException
	at net.imglib2.AbstractInterval.<init>(AbstractInterval.java:72)
	at net.imglib2.FinalInterval.<init>(FinalInterval.java:54)
	at net.imglib2.loops.LoopBuilder.<init>(LoopBuilder.java:88)
	at net.imglib2.loops.LoopBuilder.setImages(LoopBuilder.java:107)
	at com.jug.MoMA.runNetwork(MoMA.java:1885)
	at com.jug.MoMA.generateAllSimpleSegmentationHypotheses(MoMA.java:1802)
	at com.jug.MoMA.restartFromGLSegmentation(MoMA.java:2048)
	at com.jug.MoMA.processDataFromFolder(MoMA.java:1425)
	at com.jug.MoMA.main(MoMA.java:725)
Disconnected from the target VM, address: '127.0.0.1:34549', transport: 'socket'

Process finished with exit code 11

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions