Static methods in interface require -target:jvm-1.8 #265

shuDaoNan9 · 2021-03-31T08:04:33Z

I can't create any tensor, the error message is：
Error:(141, 52) Static methods in interface require -target:jvm-1.8
val fed2 = TInt32.tensorOf( org.tensorflow.ndarray.NdArrays.vectorOf(4, 3, 2, 1) )
Error:(142, 51) Static methods in interface require -target:jvm-1.8
val fed = TInt32.tensorOf(Shape.of(4)).setInt(4, 0).setInt(3, 1).setInt(2, 2).setInt(1, 3)
Error:(149, 37) Static methods in interface require -target:jvm-1.8
val itemId_index=TInt32.vectorOf(6135)
........

my scala code for test is:

    val fed2 = TInt32.tensorOf( org.tensorflow.ndarray.NdArrays.vectorOf(4, 3, 2, 1) )
    val fed = TInt32.tensorOf(Shape.of(4)).setInt(4, 0).setInt(3, 1).setInt(2, 2).setInt(1, 3)
    val a: Array[Int] =Array(1311,1205,1210,0,0,0,0,0,0,0)
    val genres = TInt32.tensorOf(StdArrays.ndCopyOf(a))
    val id_index=TInt32.vectorOf(450)

the version is:

        <dependency>
            <groupId>org.tensorflow</groupId>
            <artifactId>tensorflow-core-api</artifactId>
            <version>0.3.1</version>
        </dependency>

my environment:
Windows10 64bit,
idea 2018.3.5,
jdk1.8

Thank you!

The text was updated successfully, but these errors were encountered:

shuDaoNan9 · 2021-03-31T08:32:56Z

Infact I don't know what ${javacpp.platform.extension} is, and haven't add this:

<dependency>
  <groupId>org.tensorflow</groupId>
  <artifactId>tensorflow-core-api</artifactId>
  <version>0.3.1</version>
  <classifier>windows-x86_64${javacpp.platform.extension}</classifier>
</dependency>

Instead, I delete ${javacpp.platform.extension} like this:

  <dependency>
            <groupId>org.tensorflow</groupId>
            <artifactId>tensorflow-core-api</artifactId>
            <version>0.3.1</version>
            <classifier>windows-x86_64</classifier>
        </dependency>

Craigacp · 2021-03-31T12:22:49Z

Looks like that's a Scala issue not a TF-Java issue, you need to tell the Scala compiler to use the appropriate JVM backend - https://stackoverflow.com/questions/49213620/static-methods-in-interface-require-targetjvm-1-8.

If you use the core api jar without a specifier then it will use CPUs to execute all the code. This is fine assuming that's what you want. Otherwise you can supply one of "-mkl", "-gpu" and "-mkl-gpu" to have Intel's oneDNN support, CUDA GPU support or both.

shuDaoNan9 · 2021-04-01T02:05:07Z

have

thanks, but my scala(spark) project works well for 2 years without using TF-Java 0.3.1. And work well with the old version ( TensorFlow1.14). when I use the old version, I can create tensor like 'Tensor.create(Array(0.000231F,0.008493F))' correctly.
The old dependencies version is:

        <dependency>
            <groupId>org.tensorflow</groupId>
            <artifactId>tensorflow</artifactId>
            <version>1.14.0</version>
        </dependency>
        <dependency>
            <groupId>org.tensorflow</groupId>
            <artifactId>libtensorflow</artifactId>
            <version>1.14.0</version>
        </dependency>
        <dependency>
            <groupId>org.tensorflow</groupId>
            <artifactId>libtensorflow_jni_gpu</artifactId>
            <version>1.14.0</version>
        </dependency>
        <dependency>
            <groupId>org.tensorflow</groupId>
            <artifactId>proto</artifactId>
            <version>1.14.0</version>
        </dependency>

I will try a new scala project after that.

I did as you say, but...

Craigacp · 2021-04-01T02:22:01Z

If your Scala project wasn't calling any default methods on Java interfaces then it wouldn't have triggered the issue. We use them extensively in TF-Java, and they've been part of Java since Java 8 was released in 2014.

It looks like it found the mkl dependency, as it managed to resolve some of its dependencies. Does Maven give you an error message when building?

shuDaoNan9 · 2021-04-01T02:38:39Z

I update the last reply just now. when I try to build the project, the error message is the same:

Error:(142, 31) Static methods in interface require -target:jvm-1.8
    val fed2 = TInt32.tensorOf( org.tensorflow.ndarray.NdArrays.vectorOf(4, 3, 2, 1) )

And I can build my project correctly without try to create tensor.

I tried a new project, but facing the same error.

shuDaoNan9 · 2021-04-01T02:58:21Z

First, I trainned my model with TF2.4.0.
Second, I tried to use the PB model in my scala project with old dependencies( TF1.14 ), but failed:

Caused by: java.lang.IllegalArgumentException: ConcatOp : Expected concatenating dimensions in the range [-1, 1), but got 1
	 [[{{node StatefulPartitionedCall/StatefulPartitionedCall/model/concatenate_1/concat}}]]
	at org.tensorflow.Session.run(Native Method)
	at org.tensorflow.Session.access$100(Session.java:48)
	at org.tensorflow.Session$Runner.runHelper(Session.java:322)
	at org.tensorflow.Session$Runner.run(Session.java:272)

Third, I tried to use the PB model in my scala project with old dependencies( TF1.15 ), but failed:

java.lang.IllegalArgumentException: ConcatOp : Expected concatenating dimensions in the range [-1, 1), but got 1
	 [[{{node model/concatenate_4/concat}}]]
	at org.tensorflow.Session.run(Native Method)
	at org.tensorflow.Session.access$100(Session.java:48)
	at org.tensorflow.Session$Runner.runHelper(Session.java:326)
	at org.tensorflow.Session$Runner.run(Session.java:276)
	at xdeepfm$.getScore(xdeepfm.scala:173)

last, I tried to use the PB model in my scala project with new dependencies( TF-Java 0.3.1 ), but failed to create any tensor.

Craigacp · 2021-04-01T03:01:25Z

I update the last reply just now. when I try to build the project, the error message is the same:
Error:(142, 31) Static methods in interface require -target:jvm-1.8
    val fed2 = TInt32.tensorOf( org.tensorflow.ndarray.NdArrays.vectorOf(4, 3, 2, 1) )
And I can build my project correctly without try to create tensor.

I tried a new project, but facing the same error.

Yes, but the issue is due to your scala build, not a problem in TF-Java. The 1.x version didn't use default methods as heavily, so you didn't hit the issue. The new one does, so you need to make sure that your Scala compiler is happy with using 7 year old language features from Java 8.

Craigacp · 2021-04-01T03:02:17Z

I wouldn't expect TF 1.x to be able to load a TF 2 model so that's not too surprising.

shuDaoNan9 · 2021-04-01T05:57:25Z

not

yes, I mean this error message only appears while I try to create tensor by using TF-Java 0.3+

I update the last reply just now. when I try to build the project, the error message is the same:
Error:(142, 31) Static methods in interface require -target:jvm-1.8
    val fed2 = TInt32.tensorOf( org.tensorflow.ndarray.NdArrays.vectorOf(4, 3, 2, 1) )
And I can build my project correctly without try to create tensor.
I tried a new project, but facing the same error.
Yes, but the issue is due to your scala build, not a problem in TF-Java. The 1.x version didn't use default methods as heavily, so you didn't hit the issue. The new one does, so you need to make sure that your Scala compiler is happy with using 7 year old language features from Java 8.

I add '-target:jvm-1.8' to my setting just now:

and the error message changed, I can't load my model and create tensor now:

Warning: Could not load Loader: java.lang.UnsatisfiedLinkError: no jnijavacpp in java.library.path
Exception in thread "main" java.lang.UnsatisfiedLinkError: no jnitensorflow in java.library.path
	at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1865)
	at java.lang.Runtime.loadLibrary0(Runtime.java:870)
	at java.lang.System.loadLibrary(System.java:1122)
	at org.bytedeco.javacpp.Loader.loadLibrary(Loader.java:1683)
	at org.bytedeco.javacpp.Loader.load(Loader.java:1300)
	at org.bytedeco.javacpp.Loader.load(Loader.java:1123)
	at org.tensorflow.internal.c_api.global.tensorflow.<clinit>(tensorflow.java:12)
	at java.lang.Class.forName0(Native Method)
	at java.lang.Class.forName(Class.java:348)
	at org.bytedeco.javacpp.Loader.load(Loader.java:1190)
	at org.bytedeco.javacpp.Loader.load(Loader.java:1139)
	at org.tensorflow.NativeLibrary.load(NativeLibrary.java:48)
	at org.tensorflow.TensorFlow.init(TensorFlow.java:111)
	at org.tensorflow.TensorFlow.<clinit>(TensorFlow.java:126)
	at org.tensorflow.RawTensor.<clinit>(RawTensor.java:231)
	at org.tensorflow.Tensor.of(Tensor.java:88)
	at org.tensorflow.Tensor.of(Tensor.java:154)
	at org.tensorflow.Tensor.of(Tensor.java:126)
	at org.tensorflow.types.TInt32.tensorOf(TInt32.java:69)
...
Caused by: java.lang.UnsatisfiedLinkError: no tensorflow_cc in java.library.path
	at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1865)
	at java.lang.Runtime.loadLibrary0(Runtime.java:870)
	at java.lang.System.loadLibrary(System.java:1122)
	at org.bytedeco.javacpp.Loader.loadLibrary(Loader.java:1683)
	at org.bytedeco.javacpp.Loader.load(Loader.java:1227)
	... 16 more

So, I followed this way #259

        <dependency>
            <groupId>org.tensorflow</groupId>
            <artifactId>tensorflow-core-platform</artifactId>
            <version>0.3.1</version>
        </dependency>

Everything's all right now! Thank you very much!

shuDaoNan9 · 2021-04-01T06:09:14Z

By the way, I tried to add '-target:jvm-1.8' to my pom.xml first, but it do not work.

<build>
	<plugins>
		<plugin>
			<groupId>org.scala-tools</groupId>
			<artifactId>maven-jar-plugin</artifactId>
			<!--<version>2.3.1</version>-->
			<configuration>
				<archive>
					<manifest>
						<mainClass>my.test.classname</mainClass>
					</manifest>
				</archive>
				<addScalacArgs>-target:jvm-1.8</addScalacArgs>
			</configuration>
		</plugin>
	</plugins>
</build>

So I change my IDEA setting instead.

shuDaoNan9 closed this as completed Apr 1, 2021

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Static methods in interface require -target:jvm-1.8 #265

Static methods in interface require -target:jvm-1.8 #265

shuDaoNan9 commented Mar 31, 2021

shuDaoNan9 commented Mar 31, 2021 •

edited

Loading

Craigacp commented Mar 31, 2021 •

edited

Loading

shuDaoNan9 commented Apr 1, 2021 •

edited

Loading

Craigacp commented Apr 1, 2021

shuDaoNan9 commented Apr 1, 2021 •

edited

Loading

shuDaoNan9 commented Apr 1, 2021 •

edited

Loading

Craigacp commented Apr 1, 2021

Craigacp commented Apr 1, 2021

shuDaoNan9 commented Apr 1, 2021 •

edited

Loading

shuDaoNan9 commented Apr 1, 2021

Static methods in interface require -target:jvm-1.8 #265

Static methods in interface require -target:jvm-1.8 #265

Comments

shuDaoNan9 commented Mar 31, 2021

shuDaoNan9 commented Mar 31, 2021 • edited Loading

Craigacp commented Mar 31, 2021 • edited Loading

shuDaoNan9 commented Apr 1, 2021 • edited Loading

Craigacp commented Apr 1, 2021

shuDaoNan9 commented Apr 1, 2021 • edited Loading

shuDaoNan9 commented Apr 1, 2021 • edited Loading

Craigacp commented Apr 1, 2021

Craigacp commented Apr 1, 2021

shuDaoNan9 commented Apr 1, 2021 • edited Loading

shuDaoNan9 commented Apr 1, 2021

shuDaoNan9 commented Mar 31, 2021 •

edited

Loading

Craigacp commented Mar 31, 2021 •

edited

Loading

shuDaoNan9 commented Apr 1, 2021 •

edited

Loading

shuDaoNan9 commented Apr 1, 2021 •

edited

Loading

shuDaoNan9 commented Apr 1, 2021 •

edited

Loading

shuDaoNan9 commented Apr 1, 2021 •

edited

Loading