为什么我得到“维度 0 的切片索引 0 越界"?Java 版 Tensorflow 中的错误? [英] Why do I get "slice index 0 of dimension 0 out of bounds" error in Tensorflow for Java?
问题描述
我有一个带有以下签名的模型,我正在尝试使用 tensorflow:
MetaGraphDef with tag-set: 'serve' 包含以下 SignatureDefs:签名定义['serving_default']:给定的 SavedModel SignatureDef 包含以下输入:输入 ['jpegbase64_bytes'] 张量信息:数据类型:DT_STRING形状:(-1)名称: 占位符:0给定的 SavedModel SignatureDef 包含以下输出:输出['预测'] 张量信息:数据类型:DT_FLOAT形状:(-1, 256)名称:模型/global_average_pooling2d/Mean:0方法名称为:tensorflow/serving/predict
我调用模型的代码如下所示:
float[] predict(byte[] imageBytes) {try (Tensor result = SavedModelBundle.load("model.pb", "serve").session().runner().feed(myinput", 0, TString.tensorOfBytes(NdArrays.scalarOfObject(imageBytes))).fetch(我的输出").跑().get(0)) {浮动[]缓冲区=新的浮动[256];FloatNdArray floatNdArray = FloatDenseNdArray.create(RawDataBufferFactory.create(buffer, false),Shape.of(1, description.getNumFeatures()));((TFloat32) 结果).copyTo(floatNdArray);返回缓冲区;}}
但是,这会引发以下错误:
维度 0 的切片索引 0 越界.[[{{节点图/TensorArrayUnstack/strided_slice}}]]org.tensorflow.exceptions.TFInvalidArgumentException:维度 0 的切片索引 0 越界.[[{{节点图/TensorArrayUnstack/strided_slice}}]]在 org.tensorflow.internal.c_api.AbstractTF_Status.throwExceptionIfNotOK(AbstractTF_Status.java:87)在 org.tensorflow.Session.run(Session.java:691)在 org.tensorflow.Session.access$100(Session.java:72)在 org.tensorflow.Session$Runner.runHelper(Session.java:381)在 org.tensorflow.Session$Runner.run(Session.java:329)在 com.mridang.myapp.ImageModel.predict(ImageModel.java:69)............
据我所知,该模型需要一个密集类型的字符串张量,而我的不需要.我在 Stackoverflow 切片索引上找到了这个答案0 维 0 使用 Java API 越界 但这似乎与 张量流.
我正在使用这些依赖项:
层组:'org.tensorflow',名称:'tensorflow-core-platform',版本:'0.3.1'层组:'org.tensorflow',名称:'tensorflow-framework',版本:'0.3.1'
感谢 @jccampanero 的回答.经过一番挖掘,我在 Zoltar 库中找到了一个参考资料,展示了如何做到这一点.
https://github.com/spotify/zoltar/blob/b2c4c86f06c043aae505c533467e8a42d12da2d8/zoltar-tensorflow/src/main/java/com/spotify/zoltar/tf/TensorFlow#L6n> 我显然需要创建一个对象向量,下面的代码片段可以解决问题. I have a model with the following signature that I'm trying to invoke using tensorflow for Java: My code to invoke the model looks like this: However, this throws the following errors: From what I've understood, the model requires a dense-type string tensor while mine isn't. I found this answer on Stackoverflow slice index 0 of dimension 0 out of bounds using Java API but that seems to relate to very old version of tensorflow. I'm using these dependencies:
Thanks to @jccampanero answer. A bit of digging and I found a reference in the Zoltar library that showed how to do this. I need to create a vector of objects apparently and the following snippet did the trick.
这篇关于为什么我得到“维度 0 的切片索引 0 越界"?Java 版 Tensorflow 中的错误?的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!float[] predict(byte[] imageBytes) {try (Tensor result = SavedModelBundle.load("model.pb", "serve").session().runner().feed(myinput", 0, TString.tensorOfBytes(NdArrays.vectorOfObjects(imageBytes))).fetch(我的输出").跑().get(0)) {float[] buffer = new float[description.getNumFeatures()];FloatNdArray floatNdArray = FloatDenseNdArray.create(RawDataBufferFactory.create(buffer, false),Shape.of(1, description.getNumFeatures()));((TFloat32) 结果).copyTo(floatNdArray);返回缓冲区;}}
MetaGraphDef with tag-set: 'serve' contains the following SignatureDefs:
signature_def['serving_default']:
The given SavedModel SignatureDef contains the following input(s):
inputs['jpegbase64_bytes'] tensor_info:
dtype: DT_STRING
shape: (-1)
name: Placeholder:0
The given SavedModel SignatureDef contains the following output(s):
outputs['predictions'] tensor_info:
dtype: DT_FLOAT
shape: (-1, 256)
name: model/global_average_pooling2d/Mean:0
Method name is: tensorflow/serving/predict
float[] predict(byte[] imageBytes) {
try (Tensor result = SavedModelBundle.load("model.pb", "serve").session().runner()
.feed("myinput", 0, TString.tensorOfBytes(NdArrays.scalarOfObject(imageBytes)))
.fetch("myoutput")
.run()
.get(0)) {
float[] buffer = new float[256];
FloatNdArray floatNdArray = FloatDenseNdArray.create(RawDataBufferFactory.create(buffer, false),
Shape.of(1, description.getNumFeatures()));
((TFloat32) result).copyTo(floatNdArray);
return buffer;
}
}
slice index 0 of dimension 0 out of bounds.
[[{{node map/TensorArrayUnstack/strided_slice}}]]
org.tensorflow.exceptions.TFInvalidArgumentException: slice index 0 of dimension 0 out of bounds.
[[{{node map/TensorArrayUnstack/strided_slice}}]]
at org.tensorflow.internal.c_api.AbstractTF_Status.throwExceptionIfNotOK(AbstractTF_Status.java:87)
at org.tensorflow.Session.run(Session.java:691)
at org.tensorflow.Session.access$100(Session.java:72)
at org.tensorflow.Session$Runner.runHelper(Session.java:381)
at org.tensorflow.Session$Runner.run(Session.java:329)
at com.mridang.myapp.ImageModel.predict(ImageModel.java:69)
...
...
...
...
layer group: 'org.tensorflow', name: 'tensorflow-core-platform', version: '0.3.1'
layer group: 'org.tensorflow', name: 'tensorflow-framework', version: '0.3.1'
float[] predict(byte[] imageBytes) {
try (Tensor result = SavedModelBundle.load("model.pb", "serve").session().runner()
.feed("myinput", 0, TString.tensorOfBytes(NdArrays.vectorOfObjects(imageBytes)))
.fetch("myoutput")
.run()
.get(0)) {
float[] buffer = new float[description.getNumFeatures()];
FloatNdArray floatNdArray = FloatDenseNdArray.create(RawDataBufferFactory.create(buffer, false),
Shape.of(1, description.getNumFeatures()));
((TFloat32) result).copyTo(floatNdArray);
return buffer;
}
}