Deeplearning4j - 1.0.0-M1.1与库达-cudaGetSymbolAddress(.)失败错误

Question

我得到的错误，如标题所示。我在Stackoverflow中搜索过，其他人在以前的版本中也遇到过同样的问题。在一个回答中说，这个问题将在下一个版本的DL4J中解决，而且似乎还没有发生。下面是pom.xml和我正在使用的依赖项。拜托，有人能帮我吗？提前谢谢你。

pom.xml：

<properties>
    <dl4j-master.version>1.0.0-M1.1</dl4j-master.version>

    <logback.version>1.2.3</logback.version>
    <java.version>1.8</java.version>
    <maven-shade-plugin.version>2.4.3</maven-shade-plugin.version>
    <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
</properties>

<dependencies>

    <dependency>
        <groupId>org.deeplearning4j</groupId>
        <artifactId>deeplearning4j-core</artifactId>
        <version>${dl4j-master.version}</version>
    </dependency>

    <dependency>
        <groupId>org.deeplearning4j</groupId>
        <artifactId>deeplearning4j-nlp</artifactId>
        <version>${dl4j-master.version}</version>
    </dependency>

    <dependency>
        <groupId>org.datavec</groupId>
        <artifactId>datavec-api</artifactId>
        <version>${dl4j-master.version}</version>
    </dependency>

    <dependency>
        <groupId>org.nd4j</groupId>
        <artifactId>nd4j-cuda-11.0-platform</artifactId>
        <version>${dl4j-master.version}</version>
    </dependency>

    <dependency>
        <groupId>org.bytedeco</groupId>
        <artifactId>cuda-platform-redist</artifactId>
        <version>11.0-8.0-1.5.4</version>
    </dependency>

    <dependency>
        <groupId>org.deeplearning4j</groupId>
        <artifactId>deeplearning4j-cuda-11.0</artifactId>
        <version>${dl4j-master.version}</version>
    </dependency>

    <dependency>
        <groupId>org.bytedeco.javacpp-presets</groupId>
        <artifactId>cuda</artifactId>
        <version>10.0-7.4-1.4.4</version>
    </dependency>

    <dependency>
        <groupId>ch.qos.logback</groupId>
        <artifactId>logback-classic</artifactId>
        <version>${logback.version}</version>
    </dependency>
</dependencies>

错误：

11:11:35.720主要信息org.nd4j.linalg.api.ops.executioner.DefaultOpExecutioner -加载的JCublasBackend后端11:11:37.543主信息org.nd4j.nativeblas.NativeOpsHolder -用于线性代数的线程数: 32 11:11:37.675主信息org.nd4j.linalg.factory.Nd4jBackend -后端使用: CUDA；操作系统: Windows 10 :11:37.676主信息org.nd4j.linalg.factory.Nd4jBackend-核心: 4；内存: 3,5GB；11:11:37.676主要信息org.nd4j.linalg.api.ops.executioner.DefaultOpExecutioner - Blas供应商: CUBLAS 11:11:37.702主要信息org.nd4j.linalg.jcublas.JCublasBackend - ND4J CUDA构建版本: 11.0.221 11:11:37.705主信息org.nd4j.linalg.jcublas.JCublasBackend - CUDA设备0: NVIDIA GeForce 930 m；cc: 5.0；总内存: 4294836224 11:11:37.705主要信息org.deeplearning4j.models.sequencevectors.SequenceVectors -后端构建信息: MSVC: 192930038 STD版本:201703LCUDA: 11.0.221 DEFAULT_ENGINE: samediff::ENGINE_CUDA HAVE_FLATBUFFERS 11:11:37.782主要信息org.nd4j.linalg.jcublas.JCublasBackend -开始建立词汇表.11:11:37.783主要调试org.deeplearning4j.models.word2vec.wordstore.VocabConstructor -构建前目标词汇表大小: 11:11:37.814主调试org.deeplearning4j.models.word2vec.wordstore.VocabConstructor -尝试源迭代器: 11:11:37.814主调试org.deeplearning4j.models.word2vec.wordstore.VocabConstructor -构建前目标词汇表大小: 11:11:51.450主调试org.deeplearning4j.models.word2vec.wordstore.VocabConstructor -等待所有进程停止.11:11:51.457主要调试org.deeplearning4j.models.word2vec.wordstore.VocabConstructor -截断前的词汇表大小: 168165，NumWords: 1952392，解析序列: 318，计数器: 1952389 11:11:51.457主调试org.deeplearning4j.models.word2vec.wordstore.VocabConstructor -截断后的词汇表大小: 168165，NumWords: 1952392，解析序列: 318，计数器: 1952389 11:11:54.179主信息org.deeplearning4j.models.word2vec.wordstore.VocabConstructor -检查序列: 318，当前词汇量: 168165；序列/秒: 19,39；11:11:54.248主信息org.deeplearning4j.models.embeddings.loader.WordVectorSerializer -模型的预计内存使用:线程“java.lang.RuntimeException: cudaGetSymbolAddress(.)”中的128,30 MB异常失败；错误代码: 13在org.nd4j.linalg.jcublas.ops.executioner.CudaExecutioner.createShapeInfo(CudaExecutioner.java:2173) at org.nd4j.linalg.api.shape.Shape.createShapeInformation(Shape.java:3279) at org.nd4j.linalg.api.ndarray.BaseShapeInfoProvider.createShapeInformation(BaseShapeInfoProvider.java:75) at org.nd4j.jita.constant.ProtectedCudaShapeInfoProvider.createShapeInformation(ProtectedCudaShapeInfoProvider.java:96) at org.nd4j.jita.constant.ProtectedCudaShapeInfoProvider.createShapeInformation(ProtectedCudaShapeInfoProvider.java:77) at org.nd4j.linalg.jcublas.CachedShapeInfoProvider.createShapeInformation(org.nd4j.linalg.api.ndarray.BaseNDArray.(BaseNDArray.java:180) at org.nd4j.linalg.api.ndarray.BaseNDArray.(BaseNDArray.java:174) at org.nd4j.linalg.api.ndarray.BaseNDArray.(BaseNDArray.java:316) at org.nd4j.linalg.jcublas.JCublasNDArray.(JCublasNDArray.java:135) at org.nd4j.linalg.jcublas.JCublasNDArrayFactory.createUninitialized(JCublasNDArrayFactory.java:1533) at org.nd4j.linalg.factory.Nd4j.createUninitialized(Nd4j.java：在org.nd4j.linalg.factory.Nd4j.rand(Nd4j.java:2957) at org.nd4j.linalg.factory.Nd4j.rand(Nd4j.java:2946) at org.deeplearning4j.models.embeddings.inmemory.InMemoryLookupTable.resetWeights(InMemoryLookupTable.java:145) at org.deeplearning4j.models.sequencevectors.SequenceVectors.fit(SequenceVectors.java:278) at org.deeplearning4j.models.paragraphvectors.ParagraphVectors.fit(ParagraphVectors.java:667) at gov.rfb.cocaj.dl4jGPU.DocumentClassifier.main(DocumentClassifier.java:44)

Answer 1

这总是由于一个不兼容的cuda版本。确保您在本地安装的版本与您在dl4j中使用的版本没有什么不同。