cpu error

gistfile1.txt

Machine 2, Ubuntu 14.04, no gpu, 16 core, 32GB RAM(but free memory is about 16GB for dl4j)

java -Xmx2G -Dorg.bytedeco.javacpp.maxbytes=4G -Dorg.bytedeco.javacpp.maxphysicalbytes=4G -cp target/dl4j-examples-0.9.1-bin.jar  org.deeplearning4j.examples.recurrent.video.VideoClassificationExample
Starting data generation...
Data generation complete
o.n.l.f.Nd4jBackend - Loaded [CpuBackend] backend
o.n.n.NativeOpsHolder - Number of threads used for NativeOps: 16
o.n.n.Nd4jBlas - Number of threads used for BLAS: 16
o.n.l.a.o.e.DefaultOpExecutioner - Backend used: [CPU]; OS: [Linux]
o.n.l.a.o.e.DefaultOpExecutioner - Cores: [16]; Memory: [1.8GB];
o.n.l.a.o.e.DefaultOpExecutioner - Blas vendor: [OPENBLAS]
o.d.n.m.MultiLayerNetwork - Starting MultiLayerNetwork with WorkspaceModes set to [training: NONE; inference: SEPARATE]
Number of parameters in network: 56844
Layer 0 nParams = 9030
Layer 1 nParams = 0
Layer 2 nParams = 2710
Layer 3 nParams = 24550
Layer 4 nParams = 20350
Layer 5 nParams = 204
Starting training...
Exception in thread "main" java.lang.OutOfMemoryError: Cannot allocate new IntPointer(10): totalBytes = 76M, physicalBytes = 6G
        at org.bytedeco.javacpp.IntPointer.<init>(IntPointer.java:86)
        at org.bytedeco.javacpp.IntPointer.<init>(IntPointer.java:51)
        at org.nd4j.linalg.api.buffer.BaseDataBuffer.<init>(BaseDataBuffer.java:315)
        at org.nd4j.linalg.api.buffer.IntBuffer.<init>(IntBuffer.java:153)
        at org.nd4j.linalg.api.buffer.factory.DefaultDataBufferFactory.createInt(DefaultDataBufferFactory.java:403)
        at org.nd4j.linalg.api.buffer.factory.DefaultDataBufferFactory.createInt(DefaultDataBufferFactory.java:335)
        at org.nd4j.linalg.factory.Nd4j.createBufferDetached(Nd4j.java:1430)
        at org.nd4j.linalg.api.shape.Shape.createShapeInformation(Shape.java:2045)
        at org.nd4j.linalg.api.ndarray.BaseShapeInfoProvider.createShapeInformation(BaseShapeInfoProvider.java:47)
        at org.nd4j.linalg.cpu.nativecpu.DirectShapeInfoProvider.createShapeInformation(DirectShapeInfoProvider.java:35)
        at org.nd4j.linalg.api.ndarray.BaseNDArray.<init>(BaseNDArray.java:163)
        at org.nd4j.linalg.cpu.nativecpu.NDArray.<init>(NDArray.java:67)
        at org.nd4j.linalg.cpu.nativecpu.CpuNDArrayFactory.create(CpuNDArrayFactory.java:281)
        at org.nd4j.linalg.factory.Nd4j.create(Nd4j.java:4231)
        at org.nd4j.linalg.api.ndarray.BaseNDArray.create(BaseNDArray.java:1967)
        at org.nd4j.linalg.api.ndarray.BaseNDArray.subArray(BaseNDArray.java:2135)
        at org.nd4j.linalg.api.ndarray.BaseNDArray.get(BaseNDArray.java:4216)
        at org.deeplearning4j.nn.multilayer.MultiLayerNetwork.doTruncatedBPTT(MultiLayerNetwork.java:1441)
        at org.deeplearning4j.nn.multilayer.MultiLayerNetwork.fit(MultiLayerNetwork.java:1824)
        at org.deeplearning4j.examples.recurrent.video.VideoClassificationExample.main(VideoClassificationExample.java:157)
Caused by: java.lang.OutOfMemoryError: Physical memory usage is too high: physicalBytes = 6G > maxPhysicalBytes = 4G
        at org.bytedeco.javacpp.Pointer.deallocator(Pointer.java:576)
        at org.bytedeco.javacpp.Pointer.init(Pointer.java:121)
        at org.bytedeco.javacpp.IntPointer.allocateArray(Native Method)
        at org.bytedeco.javacpp.IntPointer.<init>(IntPointer.java:78)
        ... 19 more