Skip to content

Instantly share code, notes, and snippets.

@rkfg
Last active March 30, 2017 10:10
Show Gist options
  • Save rkfg/0d078dcd9189bf7f3d6e19b33b4979a3 to your computer and use it in GitHub Desktop.
Save rkfg/0d078dcd9189bf7f3d6e19b33b4979a3 to your computer and use it in GitHub Desktop.
13:10:07,659 INFO ~ Allocating [LOOP_ITER] workspace of 0 bytes...
13:10:07,821 INFO ~ [LOOP_ITER] spilled DEVICE array of 1152 bytes, capacity of 288 elements
13:10:07,831 INFO ~ [LOOP_ITER] spilled DEVICE array of 1152 bytes, capacity of 288 elements
13:10:07,840 INFO ~ [LOOP_ITER] spilled DEVICE array of 1152 bytes, capacity of 288 elements
13:10:07,844 WARN ~ Workspace initialization OVER_TIME was selected, but number of cycles isn't positive value!
13:10:07,844 INFO ~ Allocating [LOOP_FF] workspace of 0 bytes...
13:10:07,844 INFO ~ Allocating [LOOP_FF] workspace of 0 bytes...
13:10:07,844 INFO ~ [LOOP_FF] spilled DEVICE array of 1152 bytes, capacity of 288 elements
CUDA error at /home/rkfg/soft/svn-soft/libnd4j/blas/cuda/NativeOps.cu:4801 code=11(cudaErrorInvalidValue) "result"
Failed on [1110846756352] -> [1110844720640], size: [1152], direction: [2], result: [11]
13:10:07,846 INFO ~ Allocating [LOOP_ITER] workspace of 6912 bytes...
[WARNING]
java.lang.reflect.InvocationTargetException
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at org.codehaus.mojo.exec.ExecJavaMojo$1.run(ExecJavaMojo.java:294)
at java.lang.Thread.run(Thread.java:745)
Caused by: java.lang.IllegalStateException: MemcpyAsync failed: AllocationShape(offset=0, length=288, stride=1, elementSize=4, dataType=FLOAT)
at org.nd4j.jita.flow.impl.SynchronousFlowController.synchronizeToHost(SynchronousFlowController.java:63)
at org.nd4j.jita.flow.impl.GridFlowController.synchronizeToHost(GridFlowController.java:36)
at org.nd4j.jita.handler.impl.CudaZeroHandler.synchronizeThreadDevice(CudaZeroHandler.java:1227)
at org.nd4j.jita.allocator.impl.AtomicAllocator.synchronizeHostData(AtomicAllocator.java:321)
at org.nd4j.linalg.jcublas.buffer.BaseCudaDataBuffer.getFloat(BaseCudaDataBuffer.java:965)
at org.nd4j.linalg.jcublas.buffer.CudaFloatDataBuffer.getDouble(CudaFloatDataBuffer.java:237)
at org.nd4j.linalg.api.shape.Shape.getDouble(Shape.java:163)
at org.nd4j.linalg.api.ndarray.BaseNDArray.getDouble(BaseNDArray.java:1697)
at org.nd4j.linalg.api.ndarray.BaseNDArray.getInt(BaseNDArray.java:1666)
at org.deeplearning4j.nn.layers.feedforward.embedding.EmbeddingLayer.preOutput(EmbeddingLayer.java:93)
at org.deeplearning4j.nn.layers.feedforward.embedding.EmbeddingLayer.activate(EmbeddingLayer.java:113)
at org.deeplearning4j.nn.graph.vertex.impl.LayerVertex.doForward(LayerVertex.java:103)
at org.deeplearning4j.nn.graph.ComputationGraph.feedForward(ComputationGraph.java:1182)
at org.deeplearning4j.nn.graph.ComputationGraph.computeGradientAndScore(ComputationGraph.java:1054)
at org.deeplearning4j.optimize.solvers.BaseOptimizer.gradientAndScore(BaseOptimizer.java:160)
at org.deeplearning4j.optimize.solvers.StochasticGradientDescent.optimize(StochasticGradientDescent.java:56)
at org.deeplearning4j.optimize.Solver.optimize(Solver.java:59)
at org.deeplearning4j.nn.graph.ComputationGraph.fit(ComputationGraph.java:854)
at dlchat.EncoderDecoderLSTM.train(EncoderDecoderLSTM.java:248)
at dlchat.EncoderDecoderLSTM.run(EncoderDecoderLSTM.java:201)
at dlchat.EncoderDecoderLSTM.main(EncoderDecoderLSTM.java:149)
... 6 more
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment