tensorflow / java-models Goto Github PK

Hi, trying to run next examples, but have troubles with loading of native library.
OS: Windows 10

P.S. I see this library in the Maven Tree in IDEA

Exception in thread "main" java.lang.UnsatisfiedLinkError: no jnitensorflow in java.library.path: [C:\Program Files\JetBrains\IntelliJ IDEA Community Edition 2019.3.3\jbr\bin, C:\WINDOWS\Sun\Java\bin, C:\WINDOWS\system32, C:\WINDOWS, C:\Windows\system32, C:\Windows, C:\Windows\System32\Wbem, C:\Windows\System32\WindowsPowerShell\v1.0, C:\Windows\System32\OpenSSH, C:\Program Files\NVIDIA Corporation\NVIDIA NGX, C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common, C:\Program Files\NVIDIA Corporation\NVIDIA NvDLISR, C:\WINDOWS\system32, C:\WINDOWS, C:\WINDOWS\System32\Wbem, C:\WINDOWS\System32\WindowsPowerShell\v1.0, C:\WINDOWS\System32\OpenSSH, C:\Program Files\Git\cmd, C:\Users\zaleslaw\AppData\Local\Microsoft\WindowsApps, ., .]
at java.base/java.lang.ClassLoader.loadLibrary(ClassLoader.java:2660)
at java.base/java.lang.Runtime.loadLibrary0(Runtime.java:829)
at java.base/java.lang.System.loadLibrary(System.java:1870)
at org.bytedeco.javacpp.Loader.loadLibrary(Loader.java:1631)
at org.bytedeco.javacpp.Loader.load(Loader.java:1265)
at org.bytedeco.javacpp.Loader.load(Loader.java:1109)
at org.tensorflow.internal.c_api.global.tensorflow.(tensorflow.java:12)
at java.base/java.lang.Class.forName0(Native Method)
at java.base/java.lang.Class.forName(Class.java:398)
at org.bytedeco.javacpp.Loader.load(Loader.java:1176)
at org.bytedeco.javacpp.Loader.load(Loader.java:1125)
at org.tensorflow.NativeLibrary.load(NativeLibrary.java:48)
at org.tensorflow.TensorFlow.init(TensorFlow.java:111)
at org.tensorflow.TensorFlow.(TensorFlow.java:126)
at org.tensorflow.Graph.(Graph.java:730)
at org.tensorflow.model.examples.mnist.SimpleMnist.main(SimpleMnist.java:27)
Caused by: java.lang.UnsatisfiedLinkError: C:\Users\zaleslaw.javacpp\cache\tensorflow-core-api-0.1.0-SNAPSHOT-windows-x86_64.jar\org\tensorflow\internal\c_api\windows-x86_64\jnitensorflow.dll: Can't find dependent libraries
at java.base/java.lang.ClassLoader$NativeLibrary.load0(Native Method)
at java.base/java.lang.ClassLoader$NativeLibrary.load(ClassLoader.java:2430)
at java.base/java.lang.ClassLoader$NativeLibrary.loadLibrary(ClassLoader.java:2487)
at java.base/java.lang.ClassLoader.loadLibrary0(ClassLoader.java:2684)
at java.base/java.lang.ClassLoader.loadLibrary(ClassLoader.java:2617)
at java.base/java.lang.Runtime.load0(Runtime.java:767)
at java.base/java.lang.System.load(System.java:1834)
at org.bytedeco.javacpp.Loader.loadLibrary(Loader.java:1581)
... 12 more

Need to switch on release 0.2 and update all examples

HI :
I want to use some time series model ,I know we have // LSTMBlockCell.create()
// GRUBlockCell.create()
// BlockLSTM.create(), but I don't know how to use it,
would you like supply some example about these ?
thanks

• How to create Tensor and populate with data

• How to copy to/from tensor

• Cover basic operations on one or two tensors

• Export/import tensor to file

https://tensorflow.google.cn/api_docs/java/org/tensorflow/package-summary?authuser=1
This API about old release is deprecated and will be removed in a future version, but I can't find the API about new release.
And my TensorFlow version is 2.2.0, can I use old release(latest version is 1.15) to load model then use model to predict?

Let's generate text in William style close to this tutorial

import org.tensorflow.tools.Shape;
import org.tensorflow.tools.ndarray.ByteNdArray;
import org.tensorflow.tools.ndarray.FloatNdArray;
import org.tensorflow.tools.ndarray.index.Indices;

these imports are not found, I'd suggest the simplest solution via addition of missed dependency

In the TF1.x use the "org.tensorflow.Tensor#create(java.lang.Object)" can conveniently construct various types.
Which method in TF2 is equivalent to org.tensorflow.Tensor#create(java.lang.Object) in TF1?

There is a page here showing how to setup Tensorflow on Python for M1 ARM Macs:
https://developer.apple.com/metal/tensorflow-plugin/

It seems we could adapt this example for the M1 ARM Macs.

My question is - what is required?

Running the CNNMnist or SimpleMnist I catch the next error on my Windows 10 machine.

Looks like incorrect protobuf version in Graph creation: the exception thrown in the next row of code Graph graph = new Graph();

Exception in thread "main" java.lang.NoClassDefFoundError: com/google/protobuf/InvalidProtocolBufferException
at org.tensorflow.model.examples.mnist.CnnMnist.build(CnnMnist.java:80)
at org.tensorflow.model.examples.mnist.CnnMnist.main(CnnMnist.java:304)
Caused by: java.lang.ClassNotFoundException: com.google.protobuf.InvalidProtocolBufferException
at java.base/jdk.internal.loader.BuiltinClassLoader.loadClass(BuiltinClassLoader.java:581)
at java.base/jdk.internal.loader.ClassLoaders$AppClassLoader.loadClass(ClassLoaders.java:178)
at java.base/java.lang.ClassLoader.loadClass(ClassLoader.java:521)

• Simple Graph building

• Print out graph meta-information

• Graph serialization/deserialization

• LeNet 5 classic model on MNIST

• LeNet 5 modification with another hyper-parameters on MNIST and Fashion MNIST

• VGG'11 on FashionMnist

• ResNet on Cifar 100

• AlexNet on Cifar 100 or another subset of ImageNet

I have been using TensorFlow in Java recently to do some work. I browsed the official website and found that there are some gaps in the new TensorFlow version of the document. Then, I redirected to Github.

I am a novice, but I have seen some examples that are old version. I think we can do some updated tasks.

• Add a README file in this repo to display the functions under each folder
• tensorflow-examples module Add README file
• Is the 0.4.0 version updated to the latest Java version 0.5.0

I am happy to help improve these doc works, with the aim of quickly understanding and learning, and then sharing them with my colleague for use.