Package org.tensorflow.framework

Class CallableOptions.Builder

  • All Implemented Interfaces:
    com.google.protobuf.Message.Builder, com.google.protobuf.MessageLite.Builder, com.google.protobuf.MessageLiteOrBuilder, com.google.protobuf.MessageOrBuilder, java.lang.Cloneable, CallableOptionsOrBuilder
    Enclosing class:
    CallableOptions
    public static final class CallableOptions.Builder
extends com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>
implements CallableOptionsOrBuilder
     Defines a subgraph in another `GraphDef` as a set of feed points and nodes
 to be fetched or executed.
 Compare with the arguments to `Session::Run()`.
    Protobuf type tensorflow.CallableOptions

    • Method Detail

      • getDescriptor

        public static final com.google.protobuf.Descriptors.Descriptor getDescriptor()

      • internalGetMapField

        protected com.google.protobuf.MapField internalGetMapField​(int number)
        Overrides:
        internalGetMapField in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • internalGetMutableMapField

        protected com.google.protobuf.MapField internalGetMutableMapField​(int number)
        Overrides:
        internalGetMutableMapField in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • internalGetFieldAccessorTable

        protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable internalGetFieldAccessorTable()
        Specified by:
        internalGetFieldAccessorTable in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • clear

        public CallableOptions.Builder clear()
        Specified by:
        clear in interface com.google.protobuf.Message.Builder
        Specified by:
        clear in interface com.google.protobuf.MessageLite.Builder
        Overrides:
        clear in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • getDescriptorForType

        public com.google.protobuf.Descriptors.Descriptor getDescriptorForType()
        Specified by:
        getDescriptorForType in interface com.google.protobuf.Message.Builder
        Specified by:
        getDescriptorForType in interface com.google.protobuf.MessageOrBuilder
        Overrides:
        getDescriptorForType in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • getDefaultInstanceForType

        public CallableOptions getDefaultInstanceForType()
        Specified by:
        getDefaultInstanceForType in interface com.google.protobuf.MessageLiteOrBuilder
        Specified by:
        getDefaultInstanceForType in interface com.google.protobuf.MessageOrBuilder

      • build

        public CallableOptions build()
        Specified by:
        build in interface com.google.protobuf.Message.Builder
        Specified by:
        build in interface com.google.protobuf.MessageLite.Builder

      • buildPartial

        public CallableOptions buildPartial()
        Specified by:
        buildPartial in interface com.google.protobuf.Message.Builder
        Specified by:
        buildPartial in interface com.google.protobuf.MessageLite.Builder

      • clone

        public CallableOptions.Builder clone()
        Specified by:
        clone in interface com.google.protobuf.Message.Builder
        Specified by:
        clone in interface com.google.protobuf.MessageLite.Builder
        Overrides:
        clone in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • setField

        public CallableOptions.Builder setField​(com.google.protobuf.Descriptors.FieldDescriptor field,
                                        java.lang.Object value)
        Specified by:
        setField in interface com.google.protobuf.Message.Builder
        Overrides:
        setField in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • clearField

        public CallableOptions.Builder clearField​(com.google.protobuf.Descriptors.FieldDescriptor field)
        Specified by:
        clearField in interface com.google.protobuf.Message.Builder
        Overrides:
        clearField in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • clearOneof

        public CallableOptions.Builder clearOneof​(com.google.protobuf.Descriptors.OneofDescriptor oneof)
        Specified by:
        clearOneof in interface com.google.protobuf.Message.Builder
        Overrides:
        clearOneof in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • setRepeatedField

        public CallableOptions.Builder setRepeatedField​(com.google.protobuf.Descriptors.FieldDescriptor field,
                                                int index,
                                                java.lang.Object value)
        Specified by:
        setRepeatedField in interface com.google.protobuf.Message.Builder
        Overrides:
        setRepeatedField in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • addRepeatedField

        public CallableOptions.Builder addRepeatedField​(com.google.protobuf.Descriptors.FieldDescriptor field,
                                                java.lang.Object value)
        Specified by:
        addRepeatedField in interface com.google.protobuf.Message.Builder
        Overrides:
        addRepeatedField in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • mergeFrom

        public CallableOptions.Builder mergeFrom​(com.google.protobuf.Message other)
        Specified by:
        mergeFrom in interface com.google.protobuf.Message.Builder
        Overrides:
        mergeFrom in class com.google.protobuf.AbstractMessage.Builder<CallableOptions.Builder>

      • isInitialized

        public final boolean isInitialized()
        Specified by:
        isInitialized in interface com.google.protobuf.MessageLiteOrBuilder
        Overrides:
        isInitialized in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • mergeFrom

        public CallableOptions.Builder mergeFrom​(com.google.protobuf.CodedInputStream input,
                                         com.google.protobuf.ExtensionRegistryLite extensionRegistry)
                                  throws java.io.IOException
        Specified by:
        mergeFrom in interface com.google.protobuf.Message.Builder
        Specified by:
        mergeFrom in interface com.google.protobuf.MessageLite.Builder
        Overrides:
        mergeFrom in class com.google.protobuf.AbstractMessage.Builder<CallableOptions.Builder>
        Throws:
        java.io.IOException

      • getFeedList

        public com.google.protobuf.ProtocolStringList getFeedList()
         Tensors to be fed in the callable. Each feed is the name of a tensor.
        repeated string feed = 1;
        Specified by:
        getFeedList in interface CallableOptionsOrBuilder
        Returns:
        A list containing the feed.

      • getFeedCount

        public int getFeedCount()
         Tensors to be fed in the callable. Each feed is the name of a tensor.
        repeated string feed = 1;
        Specified by:
        getFeedCount in interface CallableOptionsOrBuilder
        Returns:
        The count of feed.

      • getFeed

        public java.lang.String getFeed​(int index)
         Tensors to be fed in the callable. Each feed is the name of a tensor.
        repeated string feed = 1;
        Specified by:
        getFeed in interface CallableOptionsOrBuilder
        Parameters:
        index - The index of the element to return.
        Returns:
        The feed at the given index.

      • getFeedBytes

        public com.google.protobuf.ByteString getFeedBytes​(int index)
         Tensors to be fed in the callable. Each feed is the name of a tensor.
        repeated string feed = 1;
        Specified by:
        getFeedBytes in interface CallableOptionsOrBuilder
        Parameters:
        index - The index of the value to return.
        Returns:
        The bytes of the feed at the given index.

      • setFeed

        public CallableOptions.Builder setFeed​(int index,
                                       java.lang.String value)
         Tensors to be fed in the callable. Each feed is the name of a tensor.
        repeated string feed = 1;
        Parameters:
        index - The index to set the value at.
        value - The feed to set.
        Returns:
        This builder for chaining.

      • addFeed

        public CallableOptions.Builder addFeed​(java.lang.String value)
         Tensors to be fed in the callable. Each feed is the name of a tensor.
        repeated string feed = 1;
        Parameters:
        value - The feed to add.
        Returns:
        This builder for chaining.

      • addAllFeed

        public CallableOptions.Builder addAllFeed​(java.lang.Iterable<java.lang.String> values)
         Tensors to be fed in the callable. Each feed is the name of a tensor.
        repeated string feed = 1;
        Parameters:
        values - The feed to add.
        Returns:
        This builder for chaining.

      • clearFeed

        public CallableOptions.Builder clearFeed()
         Tensors to be fed in the callable. Each feed is the name of a tensor.
        repeated string feed = 1;
        Returns:
        This builder for chaining.

      • addFeedBytes

        public CallableOptions.Builder addFeedBytes​(com.google.protobuf.ByteString value)
         Tensors to be fed in the callable. Each feed is the name of a tensor.
        repeated string feed = 1;
        Parameters:
        value - The bytes of the feed to add.
        Returns:
        This builder for chaining.

      • getFetchList

        public com.google.protobuf.ProtocolStringList getFetchList()
         Fetches. A list of tensor names. The caller of the callable expects a
 tensor to be returned for each fetch[i] (see RunStepResponse.tensor). The
 order of specified fetches does not change the execution order.
        repeated string fetch = 2;
        Specified by:
        getFetchList in interface CallableOptionsOrBuilder
        Returns:
        A list containing the fetch.

      • getFetchCount

        public int getFetchCount()
         Fetches. A list of tensor names. The caller of the callable expects a
 tensor to be returned for each fetch[i] (see RunStepResponse.tensor). The
 order of specified fetches does not change the execution order.
        repeated string fetch = 2;
        Specified by:
        getFetchCount in interface CallableOptionsOrBuilder
        Returns:
        The count of fetch.

      • getFetch

        public java.lang.String getFetch​(int index)
         Fetches. A list of tensor names. The caller of the callable expects a
 tensor to be returned for each fetch[i] (see RunStepResponse.tensor). The
 order of specified fetches does not change the execution order.
        repeated string fetch = 2;
        Specified by:
        getFetch in interface CallableOptionsOrBuilder
        Parameters:
        index - The index of the element to return.
        Returns:
        The fetch at the given index.

      • getFetchBytes

        public com.google.protobuf.ByteString getFetchBytes​(int index)
         Fetches. A list of tensor names. The caller of the callable expects a
 tensor to be returned for each fetch[i] (see RunStepResponse.tensor). The
 order of specified fetches does not change the execution order.
        repeated string fetch = 2;
        Specified by:
        getFetchBytes in interface CallableOptionsOrBuilder
        Parameters:
        index - The index of the value to return.
        Returns:
        The bytes of the fetch at the given index.

      • setFetch

        public CallableOptions.Builder setFetch​(int index,
                                        java.lang.String value)
         Fetches. A list of tensor names. The caller of the callable expects a
 tensor to be returned for each fetch[i] (see RunStepResponse.tensor). The
 order of specified fetches does not change the execution order.
        repeated string fetch = 2;
        Parameters:
        index - The index to set the value at.
        value - The fetch to set.
        Returns:
        This builder for chaining.

      • addFetch

        public CallableOptions.Builder addFetch​(java.lang.String value)
         Fetches. A list of tensor names. The caller of the callable expects a
 tensor to be returned for each fetch[i] (see RunStepResponse.tensor). The
 order of specified fetches does not change the execution order.
        repeated string fetch = 2;
        Parameters:
        value - The fetch to add.
        Returns:
        This builder for chaining.

      • addAllFetch

        public CallableOptions.Builder addAllFetch​(java.lang.Iterable<java.lang.String> values)
         Fetches. A list of tensor names. The caller of the callable expects a
 tensor to be returned for each fetch[i] (see RunStepResponse.tensor). The
 order of specified fetches does not change the execution order.
        repeated string fetch = 2;
        Parameters:
        values - The fetch to add.
        Returns:
        This builder for chaining.

      • clearFetch

        public CallableOptions.Builder clearFetch()
         Fetches. A list of tensor names. The caller of the callable expects a
 tensor to be returned for each fetch[i] (see RunStepResponse.tensor). The
 order of specified fetches does not change the execution order.
        repeated string fetch = 2;
        Returns:
        This builder for chaining.

      • addFetchBytes

        public CallableOptions.Builder addFetchBytes​(com.google.protobuf.ByteString value)
         Fetches. A list of tensor names. The caller of the callable expects a
 tensor to be returned for each fetch[i] (see RunStepResponse.tensor). The
 order of specified fetches does not change the execution order.
        repeated string fetch = 2;
        Parameters:
        value - The bytes of the fetch to add.
        Returns:
        This builder for chaining.

      • getTargetList

        public com.google.protobuf.ProtocolStringList getTargetList()
         Target Nodes. A list of node names. The named nodes will be run by the
 callable but their outputs will not be returned.
        repeated string target = 3;
        Specified by:
        getTargetList in interface CallableOptionsOrBuilder
        Returns:
        A list containing the target.

      • getTargetCount

        public int getTargetCount()
         Target Nodes. A list of node names. The named nodes will be run by the
 callable but their outputs will not be returned.
        repeated string target = 3;
        Specified by:
        getTargetCount in interface CallableOptionsOrBuilder
        Returns:
        The count of target.

      • getTarget

        public java.lang.String getTarget​(int index)
         Target Nodes. A list of node names. The named nodes will be run by the
 callable but their outputs will not be returned.
        repeated string target = 3;
        Specified by:
        getTarget in interface CallableOptionsOrBuilder
        Parameters:
        index - The index of the element to return.
        Returns:
        The target at the given index.

      • getTargetBytes

        public com.google.protobuf.ByteString getTargetBytes​(int index)
         Target Nodes. A list of node names. The named nodes will be run by the
 callable but their outputs will not be returned.
        repeated string target = 3;
        Specified by:
        getTargetBytes in interface CallableOptionsOrBuilder
        Parameters:
        index - The index of the value to return.
        Returns:
        The bytes of the target at the given index.

      • setTarget

        public CallableOptions.Builder setTarget​(int index,
                                         java.lang.String value)
         Target Nodes. A list of node names. The named nodes will be run by the
 callable but their outputs will not be returned.
        repeated string target = 3;
        Parameters:
        index - The index to set the value at.
        value - The target to set.
        Returns:
        This builder for chaining.

      • addTarget

        public CallableOptions.Builder addTarget​(java.lang.String value)
         Target Nodes. A list of node names. The named nodes will be run by the
 callable but their outputs will not be returned.
        repeated string target = 3;
        Parameters:
        value - The target to add.
        Returns:
        This builder for chaining.

      • addAllTarget

        public CallableOptions.Builder addAllTarget​(java.lang.Iterable<java.lang.String> values)
         Target Nodes. A list of node names. The named nodes will be run by the
 callable but their outputs will not be returned.
        repeated string target = 3;
        Parameters:
        values - The target to add.
        Returns:
        This builder for chaining.

      • clearTarget

        public CallableOptions.Builder clearTarget()
         Target Nodes. A list of node names. The named nodes will be run by the
 callable but their outputs will not be returned.
        repeated string target = 3;
        Returns:
        This builder for chaining.

      • addTargetBytes

        public CallableOptions.Builder addTargetBytes​(com.google.protobuf.ByteString value)
         Target Nodes. A list of node names. The named nodes will be run by the
 callable but their outputs will not be returned.
        repeated string target = 3;
        Parameters:
        value - The bytes of the target to add.
        Returns:
        This builder for chaining.

      • hasRunOptions

        public boolean hasRunOptions()
         Options that will be applied to each run.
        .tensorflow.RunOptions run_options = 4;
        Specified by:
        hasRunOptions in interface CallableOptionsOrBuilder
        Returns:
        Whether the runOptions field is set.

      • setRunOptions

        public CallableOptions.Builder setRunOptions​(RunOptions value)
         Options that will be applied to each run.
        .tensorflow.RunOptions run_options = 4;

      • mergeRunOptions

        public CallableOptions.Builder mergeRunOptions​(RunOptions value)
         Options that will be applied to each run.
        .tensorflow.RunOptions run_options = 4;

      • clearRunOptions

        public CallableOptions.Builder clearRunOptions()
         Options that will be applied to each run.
        .tensorflow.RunOptions run_options = 4;

      • getRunOptionsBuilder

        public RunOptions.Builder getRunOptionsBuilder()
         Options that will be applied to each run.
        .tensorflow.RunOptions run_options = 4;

      • getTensorConnectionList

        public java.util.List<TensorConnection> getTensorConnectionList()
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;
        Specified by:
        getTensorConnectionList in interface CallableOptionsOrBuilder

      • getTensorConnectionCount

        public int getTensorConnectionCount()
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;
        Specified by:
        getTensorConnectionCount in interface CallableOptionsOrBuilder

      • getTensorConnection

        public TensorConnection getTensorConnection​(int index)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;
        Specified by:
        getTensorConnection in interface CallableOptionsOrBuilder

      • setTensorConnection

        public CallableOptions.Builder setTensorConnection​(int index,
                                                   TensorConnection value)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • setTensorConnection

        public CallableOptions.Builder setTensorConnection​(int index,
                                                   TensorConnection.Builder builderForValue)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • addTensorConnection

        public CallableOptions.Builder addTensorConnection​(TensorConnection value)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • addTensorConnection

        public CallableOptions.Builder addTensorConnection​(int index,
                                                   TensorConnection value)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • addTensorConnection

        public CallableOptions.Builder addTensorConnection​(TensorConnection.Builder builderForValue)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • addTensorConnection

        public CallableOptions.Builder addTensorConnection​(int index,
                                                   TensorConnection.Builder builderForValue)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • addAllTensorConnection

        public CallableOptions.Builder addAllTensorConnection​(java.lang.Iterable<? extends TensorConnection> values)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • clearTensorConnection

        public CallableOptions.Builder clearTensorConnection()
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • removeTensorConnection

        public CallableOptions.Builder removeTensorConnection​(int index)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • getTensorConnectionBuilder

        public TensorConnection.Builder getTensorConnectionBuilder​(int index)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • getTensorConnectionOrBuilder

        public TensorConnectionOrBuilder getTensorConnectionOrBuilder​(int index)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;
        Specified by:
        getTensorConnectionOrBuilder in interface CallableOptionsOrBuilder

      • getTensorConnectionOrBuilderList

        public java.util.List<? extends TensorConnectionOrBuilder> getTensorConnectionOrBuilderList()
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;
        Specified by:
        getTensorConnectionOrBuilderList in interface CallableOptionsOrBuilder

      • addTensorConnectionBuilder

        public TensorConnection.Builder addTensorConnectionBuilder()
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • addTensorConnectionBuilder

        public TensorConnection.Builder addTensorConnectionBuilder​(int index)
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • getTensorConnectionBuilderList

        public java.util.List<TensorConnection.Builder> getTensorConnectionBuilderList()
         Tensors to be connected in the callable. Each TensorConnection denotes
 a pair of tensors in the graph, between which an edge will be created
 in the callable.
        repeated .tensorflow.TensorConnection tensor_connection = 5;

      • getFeedDevicesCount

        public int getFeedDevicesCount()
        Description copied from interface: CallableOptionsOrBuilder
         The Tensor objects fed in the callable and fetched from the callable
 are expected to be backed by host (CPU) memory by default.
 The options below allow changing that - feeding tensors backed by
 device memory, or returning tensors that are backed by device memory.
 The maps below map the name of a feed/fetch tensor (which appears in
 'feed' or 'fetch' fields above), to the fully qualified name of the device
 owning the memory backing the contents of the tensor.
 For example, creating a callable with the following options:
 CallableOptions {
   feed: "a:0"
   feed: "b:0"
   fetch: "x:0"
   fetch: "y:0"
   feed_devices: {
     "a:0": "/job:localhost/replica:0/task:0/device:GPU:0"
   }
   fetch_devices: {
     "y:0": "/job:localhost/replica:0/task:0/device:GPU:0"
  }
 }
 means that the Callable expects:
 - The first argument ("a:0") is a Tensor backed by GPU memory.
 - The second argument ("b:0") is a Tensor backed by host memory.
 and of its return values:
 - The first output ("x:0") will be backed by host memory.
 - The second output ("y:0") will be backed by GPU memory.
 FEEDS:
 It is the responsibility of the caller to ensure that the memory of the fed
 tensors will be correctly initialized and synchronized before it is
 accessed by operations executed during the call to Session::RunCallable().
 This is typically ensured by using the TensorFlow memory allocators
 (Device::GetAllocator()) to create the Tensor to be fed.
 Alternatively, for CUDA-enabled GPU devices, this typically means that the
 operation that produced the contents of the tensor has completed, i.e., the
 CUDA stream has been synchronized (e.g., via cuCtxSynchronize() or
 cuStreamSynchronize()).
        map<string, string> feed_devices = 6;
        Specified by:
        getFeedDevicesCount in interface CallableOptionsOrBuilder

      • containsFeedDevices

        public boolean containsFeedDevices​(java.lang.String key)
         The Tensor objects fed in the callable and fetched from the callable
 are expected to be backed by host (CPU) memory by default.
 The options below allow changing that - feeding tensors backed by
 device memory, or returning tensors that are backed by device memory.
 The maps below map the name of a feed/fetch tensor (which appears in
 'feed' or 'fetch' fields above), to the fully qualified name of the device
 owning the memory backing the contents of the tensor.
 For example, creating a callable with the following options:
 CallableOptions {
   feed: "a:0"
   feed: "b:0"
   fetch: "x:0"
   fetch: "y:0"
   feed_devices: {
     "a:0": "/job:localhost/replica:0/task:0/device:GPU:0"
   }
   fetch_devices: {
     "y:0": "/job:localhost/replica:0/task:0/device:GPU:0"
  }
 }
 means that the Callable expects:
 - The first argument ("a:0") is a Tensor backed by GPU memory.
 - The second argument ("b:0") is a Tensor backed by host memory.
 and of its return values:
 - The first output ("x:0") will be backed by host memory.
 - The second output ("y:0") will be backed by GPU memory.
 FEEDS:
 It is the responsibility of the caller to ensure that the memory of the fed
 tensors will be correctly initialized and synchronized before it is
 accessed by operations executed during the call to Session::RunCallable().
 This is typically ensured by using the TensorFlow memory allocators
 (Device::GetAllocator()) to create the Tensor to be fed.
 Alternatively, for CUDA-enabled GPU devices, this typically means that the
 operation that produced the contents of the tensor has completed, i.e., the
 CUDA stream has been synchronized (e.g., via cuCtxSynchronize() or
 cuStreamSynchronize()).
        map<string, string> feed_devices = 6;
        Specified by:
        containsFeedDevices in interface CallableOptionsOrBuilder

      • getFeedDevicesMap

        public java.util.Map<java.lang.String,​java.lang.String> getFeedDevicesMap()
         The Tensor objects fed in the callable and fetched from the callable
 are expected to be backed by host (CPU) memory by default.
 The options below allow changing that - feeding tensors backed by
 device memory, or returning tensors that are backed by device memory.
 The maps below map the name of a feed/fetch tensor (which appears in
 'feed' or 'fetch' fields above), to the fully qualified name of the device
 owning the memory backing the contents of the tensor.
 For example, creating a callable with the following options:
 CallableOptions {
   feed: "a:0"
   feed: "b:0"
   fetch: "x:0"
   fetch: "y:0"
   feed_devices: {
     "a:0": "/job:localhost/replica:0/task:0/device:GPU:0"
   }
   fetch_devices: {
     "y:0": "/job:localhost/replica:0/task:0/device:GPU:0"
  }
 }
 means that the Callable expects:
 - The first argument ("a:0") is a Tensor backed by GPU memory.
 - The second argument ("b:0") is a Tensor backed by host memory.
 and of its return values:
 - The first output ("x:0") will be backed by host memory.
 - The second output ("y:0") will be backed by GPU memory.
 FEEDS:
 It is the responsibility of the caller to ensure that the memory of the fed
 tensors will be correctly initialized and synchronized before it is
 accessed by operations executed during the call to Session::RunCallable().
 This is typically ensured by using the TensorFlow memory allocators
 (Device::GetAllocator()) to create the Tensor to be fed.
 Alternatively, for CUDA-enabled GPU devices, this typically means that the
 operation that produced the contents of the tensor has completed, i.e., the
 CUDA stream has been synchronized (e.g., via cuCtxSynchronize() or
 cuStreamSynchronize()).
        map<string, string> feed_devices = 6;
        Specified by:
        getFeedDevicesMap in interface CallableOptionsOrBuilder

      • getFeedDevicesOrDefault

        public java.lang.String getFeedDevicesOrDefault​(java.lang.String key,
                                                java.lang.String defaultValue)
         The Tensor objects fed in the callable and fetched from the callable
 are expected to be backed by host (CPU) memory by default.
 The options below allow changing that - feeding tensors backed by
 device memory, or returning tensors that are backed by device memory.
 The maps below map the name of a feed/fetch tensor (which appears in
 'feed' or 'fetch' fields above), to the fully qualified name of the device
 owning the memory backing the contents of the tensor.
 For example, creating a callable with the following options:
 CallableOptions {
   feed: "a:0"
   feed: "b:0"
   fetch: "x:0"
   fetch: "y:0"
   feed_devices: {
     "a:0": "/job:localhost/replica:0/task:0/device:GPU:0"
   }
   fetch_devices: {
     "y:0": "/job:localhost/replica:0/task:0/device:GPU:0"
  }
 }
 means that the Callable expects:
 - The first argument ("a:0") is a Tensor backed by GPU memory.
 - The second argument ("b:0") is a Tensor backed by host memory.
 and of its return values:
 - The first output ("x:0") will be backed by host memory.
 - The second output ("y:0") will be backed by GPU memory.
 FEEDS:
 It is the responsibility of the caller to ensure that the memory of the fed
 tensors will be correctly initialized and synchronized before it is
 accessed by operations executed during the call to Session::RunCallable().
 This is typically ensured by using the TensorFlow memory allocators
 (Device::GetAllocator()) to create the Tensor to be fed.
 Alternatively, for CUDA-enabled GPU devices, this typically means that the
 operation that produced the contents of the tensor has completed, i.e., the
 CUDA stream has been synchronized (e.g., via cuCtxSynchronize() or
 cuStreamSynchronize()).
        map<string, string> feed_devices = 6;
        Specified by:
        getFeedDevicesOrDefault in interface CallableOptionsOrBuilder

      • getFeedDevicesOrThrow

        public java.lang.String getFeedDevicesOrThrow​(java.lang.String key)
         The Tensor objects fed in the callable and fetched from the callable
 are expected to be backed by host (CPU) memory by default.
 The options below allow changing that - feeding tensors backed by
 device memory, or returning tensors that are backed by device memory.
 The maps below map the name of a feed/fetch tensor (which appears in
 'feed' or 'fetch' fields above), to the fully qualified name of the device
 owning the memory backing the contents of the tensor.
 For example, creating a callable with the following options:
 CallableOptions {
   feed: "a:0"
   feed: "b:0"
   fetch: "x:0"
   fetch: "y:0"
   feed_devices: {
     "a:0": "/job:localhost/replica:0/task:0/device:GPU:0"
   }
   fetch_devices: {
     "y:0": "/job:localhost/replica:0/task:0/device:GPU:0"
  }
 }
 means that the Callable expects:
 - The first argument ("a:0") is a Tensor backed by GPU memory.
 - The second argument ("b:0") is a Tensor backed by host memory.
 and of its return values:
 - The first output ("x:0") will be backed by host memory.
 - The second output ("y:0") will be backed by GPU memory.
 FEEDS:
 It is the responsibility of the caller to ensure that the memory of the fed
 tensors will be correctly initialized and synchronized before it is
 accessed by operations executed during the call to Session::RunCallable().
 This is typically ensured by using the TensorFlow memory allocators
 (Device::GetAllocator()) to create the Tensor to be fed.
 Alternatively, for CUDA-enabled GPU devices, this typically means that the
 operation that produced the contents of the tensor has completed, i.e., the
 CUDA stream has been synchronized (e.g., via cuCtxSynchronize() or
 cuStreamSynchronize()).
        map<string, string> feed_devices = 6;
        Specified by:
        getFeedDevicesOrThrow in interface CallableOptionsOrBuilder

      • removeFeedDevices

        public CallableOptions.Builder removeFeedDevices​(java.lang.String key)
         The Tensor objects fed in the callable and fetched from the callable
 are expected to be backed by host (CPU) memory by default.
 The options below allow changing that - feeding tensors backed by
 device memory, or returning tensors that are backed by device memory.
 The maps below map the name of a feed/fetch tensor (which appears in
 'feed' or 'fetch' fields above), to the fully qualified name of the device
 owning the memory backing the contents of the tensor.
 For example, creating a callable with the following options:
 CallableOptions {
   feed: "a:0"
   feed: "b:0"
   fetch: "x:0"
   fetch: "y:0"
   feed_devices: {
     "a:0": "/job:localhost/replica:0/task:0/device:GPU:0"
   }
   fetch_devices: {
     "y:0": "/job:localhost/replica:0/task:0/device:GPU:0"
  }
 }
 means that the Callable expects:
 - The first argument ("a:0") is a Tensor backed by GPU memory.
 - The second argument ("b:0") is a Tensor backed by host memory.
 and of its return values:
 - The first output ("x:0") will be backed by host memory.
 - The second output ("y:0") will be backed by GPU memory.
 FEEDS:
 It is the responsibility of the caller to ensure that the memory of the fed
 tensors will be correctly initialized and synchronized before it is
 accessed by operations executed during the call to Session::RunCallable().
 This is typically ensured by using the TensorFlow memory allocators
 (Device::GetAllocator()) to create the Tensor to be fed.
 Alternatively, for CUDA-enabled GPU devices, this typically means that the
 operation that produced the contents of the tensor has completed, i.e., the
 CUDA stream has been synchronized (e.g., via cuCtxSynchronize() or
 cuStreamSynchronize()).
        map<string, string> feed_devices = 6;

      • getMutableFeedDevices

        @Deprecated
public java.util.Map<java.lang.String,​java.lang.String> getMutableFeedDevices()
        Deprecated.
        Use alternate mutation accessors instead.

      • putFeedDevices

        public CallableOptions.Builder putFeedDevices​(java.lang.String key,
                                              java.lang.String value)
         The Tensor objects fed in the callable and fetched from the callable
 are expected to be backed by host (CPU) memory by default.
 The options below allow changing that - feeding tensors backed by
 device memory, or returning tensors that are backed by device memory.
 The maps below map the name of a feed/fetch tensor (which appears in
 'feed' or 'fetch' fields above), to the fully qualified name of the device
 owning the memory backing the contents of the tensor.
 For example, creating a callable with the following options:
 CallableOptions {
   feed: "a:0"
   feed: "b:0"
   fetch: "x:0"
   fetch: "y:0"
   feed_devices: {
     "a:0": "/job:localhost/replica:0/task:0/device:GPU:0"
   }
   fetch_devices: {
     "y:0": "/job:localhost/replica:0/task:0/device:GPU:0"
  }
 }
 means that the Callable expects:
 - The first argument ("a:0") is a Tensor backed by GPU memory.
 - The second argument ("b:0") is a Tensor backed by host memory.
 and of its return values:
 - The first output ("x:0") will be backed by host memory.
 - The second output ("y:0") will be backed by GPU memory.
 FEEDS:
 It is the responsibility of the caller to ensure that the memory of the fed
 tensors will be correctly initialized and synchronized before it is
 accessed by operations executed during the call to Session::RunCallable().
 This is typically ensured by using the TensorFlow memory allocators
 (Device::GetAllocator()) to create the Tensor to be fed.
 Alternatively, for CUDA-enabled GPU devices, this typically means that the
 operation that produced the contents of the tensor has completed, i.e., the
 CUDA stream has been synchronized (e.g., via cuCtxSynchronize() or
 cuStreamSynchronize()).
        map<string, string> feed_devices = 6;

      • putAllFeedDevices

        public CallableOptions.Builder putAllFeedDevices​(java.util.Map<java.lang.String,​java.lang.String> values)
         The Tensor objects fed in the callable and fetched from the callable
 are expected to be backed by host (CPU) memory by default.
 The options below allow changing that - feeding tensors backed by
 device memory, or returning tensors that are backed by device memory.
 The maps below map the name of a feed/fetch tensor (which appears in
 'feed' or 'fetch' fields above), to the fully qualified name of the device
 owning the memory backing the contents of the tensor.
 For example, creating a callable with the following options:
 CallableOptions {
   feed: "a:0"
   feed: "b:0"
   fetch: "x:0"
   fetch: "y:0"
   feed_devices: {
     "a:0": "/job:localhost/replica:0/task:0/device:GPU:0"
   }
   fetch_devices: {
     "y:0": "/job:localhost/replica:0/task:0/device:GPU:0"
  }
 }
 means that the Callable expects:
 - The first argument ("a:0") is a Tensor backed by GPU memory.
 - The second argument ("b:0") is a Tensor backed by host memory.
 and of its return values:
 - The first output ("x:0") will be backed by host memory.
 - The second output ("y:0") will be backed by GPU memory.
 FEEDS:
 It is the responsibility of the caller to ensure that the memory of the fed
 tensors will be correctly initialized and synchronized before it is
 accessed by operations executed during the call to Session::RunCallable().
 This is typically ensured by using the TensorFlow memory allocators
 (Device::GetAllocator()) to create the Tensor to be fed.
 Alternatively, for CUDA-enabled GPU devices, this typically means that the
 operation that produced the contents of the tensor has completed, i.e., the
 CUDA stream has been synchronized (e.g., via cuCtxSynchronize() or
 cuStreamSynchronize()).
        map<string, string> feed_devices = 6;

      • getFetchDevicesMap

        public java.util.Map<java.lang.String,​java.lang.String> getFetchDevicesMap()
        map<string, string> fetch_devices = 7;
        Specified by:
        getFetchDevicesMap in interface CallableOptionsOrBuilder

      • getFetchDevicesOrDefault

        public java.lang.String getFetchDevicesOrDefault​(java.lang.String key,
                                                 java.lang.String defaultValue)
        map<string, string> fetch_devices = 7;
        Specified by:
        getFetchDevicesOrDefault in interface CallableOptionsOrBuilder

      • removeFetchDevices

        public CallableOptions.Builder removeFetchDevices​(java.lang.String key)
        map<string, string> fetch_devices = 7;

      • getMutableFetchDevices

        @Deprecated
public java.util.Map<java.lang.String,​java.lang.String> getMutableFetchDevices()
        Deprecated.
        Use alternate mutation accessors instead.

      • putFetchDevices

        public CallableOptions.Builder putFetchDevices​(java.lang.String key,
                                               java.lang.String value)
        map<string, string> fetch_devices = 7;

      • putAllFetchDevices

        public CallableOptions.Builder putAllFetchDevices​(java.util.Map<java.lang.String,​java.lang.String> values)
        map<string, string> fetch_devices = 7;

      • getFetchSkipSync

        public boolean getFetchSkipSync()
         By default, RunCallable() will synchronize the GPU stream before returning
 fetched tensors on a GPU device, to ensure that the values in those tensors
 have been produced. This simplifies interacting with the tensors, but
 potentially incurs a performance hit.
 If this options is set to true, the caller is responsible for ensuring
 that the values in the fetched tensors have been produced before they are
 used. The caller can do this by invoking `Device::Sync()` on the underlying
 device(s), or by feeding the tensors back to the same Session using
 `feed_devices` with the same corresponding device name.
        bool fetch_skip_sync = 8;
        Specified by:
        getFetchSkipSync in interface CallableOptionsOrBuilder
        Returns:
        The fetchSkipSync.

      • setFetchSkipSync

        public CallableOptions.Builder setFetchSkipSync​(boolean value)
         By default, RunCallable() will synchronize the GPU stream before returning
 fetched tensors on a GPU device, to ensure that the values in those tensors
 have been produced. This simplifies interacting with the tensors, but
 potentially incurs a performance hit.
 If this options is set to true, the caller is responsible for ensuring
 that the values in the fetched tensors have been produced before they are
 used. The caller can do this by invoking `Device::Sync()` on the underlying
 device(s), or by feeding the tensors back to the same Session using
 `feed_devices` with the same corresponding device name.
        bool fetch_skip_sync = 8;
        Parameters:
        value - The fetchSkipSync to set.
        Returns:
        This builder for chaining.

      • clearFetchSkipSync

        public CallableOptions.Builder clearFetchSkipSync()
         By default, RunCallable() will synchronize the GPU stream before returning
 fetched tensors on a GPU device, to ensure that the values in those tensors
 have been produced. This simplifies interacting with the tensors, but
 potentially incurs a performance hit.
 If this options is set to true, the caller is responsible for ensuring
 that the values in the fetched tensors have been produced before they are
 used. The caller can do this by invoking `Device::Sync()` on the underlying
 device(s), or by feeding the tensors back to the same Session using
 `feed_devices` with the same corresponding device name.
        bool fetch_skip_sync = 8;
        Returns:
        This builder for chaining.

      • setUnknownFields

        public final CallableOptions.Builder setUnknownFields​(com.google.protobuf.UnknownFieldSet unknownFields)
        Specified by:
        setUnknownFields in interface com.google.protobuf.Message.Builder
        Overrides:
        setUnknownFields in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>

      • mergeUnknownFields

        public final CallableOptions.Builder mergeUnknownFields​(com.google.protobuf.UnknownFieldSet unknownFields)
        Specified by:
        mergeUnknownFields in interface com.google.protobuf.Message.Builder
        Overrides:
        mergeUnknownFields in class com.google.protobuf.GeneratedMessageV3.Builder<CallableOptions.Builder>