InferenceOutput.Builder
public
static
final
class
InferenceOutput.Builder
extends Object
java.lang.Object | |
↳ | android.adservices.ondevicepersonalization.InferenceOutput.Builder |
A builder for InferenceOutput
Summary
Public constructors | |
---|---|
Builder()
|
Public methods | |
---|---|
InferenceOutput.Builder
|
addDataOutput(int key, Object value)
|
InferenceOutput
|
build()
Builds the instance. |
InferenceOutput.Builder
|
setData(byte[] value)
A byte array that holds input data. |
InferenceOutput.Builder
|
setDataOutputs(Map<Integer, Object> value)
Note: use |
Inherited methods | |
---|---|
Public constructors
Public methods
addDataOutput
public InferenceOutput.Builder addDataOutput (int key, Object value)
Parameters | |
---|---|
key |
int |
value |
Object : This value cannot be null . |
Returns | |
---|---|
InferenceOutput.Builder |
This value cannot be null . |
See also:
build
public InferenceOutput build ()
Builds the instance. This builder should not be touched after calling this!
Returns | |
---|---|
InferenceOutput |
This value cannot be null . |
setData
public InferenceOutput.Builder setData (byte[] value)
A byte array that holds input data. The inputs should be in the same order as inputs of the model.
For LiteRT, this field is a serialized Map For ExecuTorch model, this field is a serialized EValue array.
Parameters | |
---|---|
value |
byte : This value cannot be null . |
Returns | |
---|---|
InferenceOutput.Builder |
This value cannot be null . |
setDataOutputs
public InferenceOutput.Builder setDataOutputs (Map<Integer, Object> value)
Note: use InferenceOutput.Builder.setData(byte[])
instead.
A map mapping output indices to multidimensional arrays of output.
For TFLite, this field is mapped to outputs of runForMultipleInputsOutputs: https://www.tensorflow.org/lite/api_docs/java/org/tensorflow/lite/InterpreterApi#parameters_9
Parameters | |
---|---|
value |
Map : This value cannot be null . |
Returns | |
---|---|
InferenceOutput.Builder |
This value cannot be null . |