NLClassifierOptions class
Options to identify input and output tensors of the model.
Configure the input/output tensors for NLClassifier:
- No special configuration is needed if the model has only one input tensor and one output tensor.
- When the model has multiple input or output tensors, use the following configurations
to specifiy the desired tensors:
-- tensor names: {inputTensorName}, {outputScoreTensorName}, {@code
outputLabelTensorName}
-- tensor indices: {inputTensorIndex}, {outputScoreTensorIndex}, {@code
outputLabelTensorIndex}
Tensor names has higher priorities than tensor indices in locating the tensors. It means
the tensors will be first located according to tensor names. If not found, then the tensors
will be located according to tensor indices.
- Failing to match the input text tensor or output score tensor with neither tensor names nor tensor indices will trigger a runtime error. However, failing to locate the output label tensor will not trigger an error because the label tensor is optional.
Constructors
- NLClassifierOptions()
-
Creates a new options instance.
factory
Properties
-
base
→ Pointer<
TfLiteNLClassifierOptions> -
no setter
- hashCode → int
-
The hash code for this object.
no setterinherited
- inputTensorIndex ↔ int
-
Set the index of the input text tensor among all input tensors, if the model has multiple
inputs. Only the input tensor specified will be used for inference; other input tensors
will be ignored. Dafualt to 0.
getter/setter pair
- inputTensorName ↔ String
-
Set the name of the input text tensor, if the model has multiple inputs. Only the input
tensor specified will be used for inference; other input tensors will be ignored. Dafualt
to
"INPUT"
.getter/setter pair - outputLabelTensorIndex ↔ int
-
Set the index of the optional output label tensor among all output tensors, if the model
has multiple outputs.
getter/setter pair
- outputLabelTensorName ↔ String
-
Set the name of the output label tensor, if the model has multiple outputs. Dafualt to
"OUTPUT_LABEL"
.getter/setter pair - outputScoreTensorIndex ↔ int
-
Set the index of the output score tensor among all output tensors, if the model has
multiple outputs. Dafualt to 0.
getter/setter pair
- outputScoreTensorName ↔ String
-
Set the name of the output score tensor, if the model has multiple outputs. Dafualt to
"OUTPUT_SCORE"
.getter/setter pair - runtimeType → Type
-
A representation of the runtime type of the object.
no setterinherited
Methods
-
delete(
) → void - Destroys the options instance.
-
noSuchMethod(
Invocation invocation) → dynamic -
Invoked when a nonexistent method or property is accessed.
inherited
-
toString(
) → String -
A string representation of this object.
inherited
Operators
-
operator ==(
Object other) → bool -
The equality operator.
inherited
Constants
- DEFAULT_INPUT_TENSOR_INDEX → const int
- DEFAULT_INPUT_TENSOR_NAME → const String
- DEFAULT_OUTPUT_LABEL_TENSOR_INDEX → const int
- DEFAULT_OUTPUT_LABEL_TENSOR_NAME → const String
- DEFAULT_OUTPUT_SCORE_TENSOR_INDEX → const int
- DEFAULT_OUTPUT_SCORE_TENSOR_NAME → const String