Introducing LiteRT Next: A new set of APIs that improves and simplifies on-device hardware acceleration.

TensorBufferRequirements

class TensorBufferRequirements : JniHandle

kotlin.Any
↳	com.google.ai.edge.litert.JniHandle
	↳	com.google.ai.edge.litert.TensorBufferRequirements

Requirements for allocating a TensorBuffer.

Summary

Public functions
`Int`	`@<Error class: unknown class> bufferSize()` Returns the size of the tensor buffer requirements in bytes.
`IntArray`	`@<Error class: unknown class> strides()` Returns the strides of the tensor buffer requirements.
`List<TensorBufferType>`	`@<Error class: unknown class> supportedTypes()` Returns all the types supported by the tensor buffer requirements.

Protected functions
`open Unit`	`destroy()` Clean up resources associated with the handle.

`Unit`	`assertNotDestroyed()` Asserts that the handle is not destroyed, otherwise throws an `IllegalStateException`.
`final Unit`	`close()` Clean up the handle safely to avoid releasing the same JNI handle multiple times.

@<Error class: unknown class>
fun bufferSize(): Int

Returns the size of the tensor buffer requirements in bytes.

@<Error class: unknown class>
fun strides(): IntArray

Returns the strides of the tensor buffer requirements.

@<Error class: unknown class>
fun supportedTypes(): List<TensorBufferType>

Returns all the types supported by the tensor buffer requirements.

protected open fun destroy(): Unit

Clean up resources associated with the handle.