Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit ad2db12b authored by David Gross's avatar David Gross
Browse files

Documentation improvements. 

Document that every model must have at least one input and one output.

Document graph constraints.

Elaborate a bit on "execution order".

Indicate how operands of different lifetimes are defined (or not).

Fix ANEURALNETWORKS_L2_NORMALIZATION documentation for output shape.

ANEURALNETWORKS_SPACE_TO_DEPTH has inconsistent names for dimensions.

ANEURALNETWORKS_PAD documentation doesn't clearly describe rank or
dimensions of output tensor.

Bug: 66478689
Bug: 67601776
Bug: 69631404
Bug: 79482779
Bug: 79539157
Bug: 79539928

Test: mma at tree root

Change-Id: I6227bea2a486d4b45022d590915e10a124d323d4
Merged-In: I6227bea2a486d4b45022d590915e10a124d323d4
parent 2a970cb8

current.txt

+1 −1
Original line number Diff line number Diff line
@@ -250,7 +250,7 @@ c8bc853546dd55584611def2a9fa1d99f657e3366c976d2f60fe6b8aa6d2cb87 android.hardwar 
5804ca86611d72e5481f022b3a0c1b334217f2e4988dad25730c42af2d1f4d1c android.hardware.neuralnetworks@1.0::IDevice

12e8dca4ab7d8aadd0ef8f1b438021938e2396139e85db2ed65783b08800aa52 android.hardware.neuralnetworks@1.0::IExecutionCallback

574e8f1499436fb4075894dcae0b36682427956ecb114f17f1fe22d116a83c6b android.hardware.neuralnetworks@1.0::IPreparedModel

b07802a3433a9abec67a41525dc04eab9efb62760414e1d37e2f72cc5ae9cb8a android.hardware.neuralnetworks@1.0::types

1a5ae9793223658174258b523763c557abad6fb917df0b8e3cc097fc89035811 android.hardware.neuralnetworks@1.0::types



# Documentation fixups for b/78135149

9e7a0b650d0e461ece2cfec0e1072abf8676f592b41a7fb48f01e88fc3c8f780 android.hardware.broadcastradio@1.0::types

neuralnetworks/1.0/types.hal

+23 −13
Original line number Diff line number Diff line
@@ -596,8 +596,8 @@ enum OperationType : int32_t { 
     * * 0: A 4-D tensor, of shape [batches, height, width, depth].

     *

     * Outputs:

     * * 0: The output 4-D tensor, of shape

     *      [batches, out_height, out_width, depth].

     * * 0: The output 4-D tensor, of the same shape as input

     *      [batches, height, width, depth].

     */

    L2_NORMALIZATION = 11,
@@ -1282,8 +1282,8 @@ enum OperationType : int32_t { 
     *      input height and width.

     *

     * Outputs:

     * * 0: The output 4-D tensor, of shape [batch, height/block_size,

     *      width/block_size, depth*block_size*block_size].

     * * 0: The output 4-D tensor, of shape [batches, height/block_size,

     *      width/block_size, depth_in*block_size*block_size].

     */

    SPACE_TO_DEPTH = 26,
@@ -1410,28 +1410,36 @@ enum FusedActivationFunc : int32_t { 
enum OperandLifeTime : int32_t {

    /**

     * The operand is internal to the model. It's created by an operation and

     * consumed by other operations.

     * consumed by other operations. It must be an output operand of

     * exactly one operation.

     */

    TEMPORARY_VARIABLE,



    /**

     * The operand is an input of the model. An operand can't be both

     * input and output of a model.

     * The operand is an input of the model. It must not be an output

     * operand of any operation.

     *

     * An operand can't be both input and output of a model.

     */

    MODEL_INPUT,



    /**

     * The operand is an output of the model.

     * The operand is an output of the model. It must be an output

     * operand of exactly one operation.

     *

     * An operand can't be both input and output of a model.

     */

    MODEL_OUTPUT,



    /**

     * The operand is a constant found in Model.operandValues.

     * The operand is a constant found in Model.operandValues. It must

     * not be an output operand of any operation.

     */

    CONSTANT_COPY,



    /**

     * The operand is a constant that was specified via a Memory object.

     * The operand is a constant that was specified via a Memory

     * object. It must not be an output operand of any operation.

     */

    CONSTANT_REFERENCE,
@@ -1641,19 +1649,21 @@ struct Model { 
    /**

     * All operations included in the model.

     *

     * The operations are sorted into execution order.

     * The operations are sorted into execution order. Every operand

     * with lifetime MODEL_OUTPUT or TEMPORARY_VARIABLE must be

     * written before it is read.

     */

    vec<Operation> operations;



    /**

     * Input indexes of the model.

     * Input indexes of the model. There must be at least one.

     *

     * Each value corresponds to the index of the operand in "operands".

     */

    vec<uint32_t> inputIndexes;



    /**

     * Output indexes of the model.

     * Output indexes of the model. There must be at least one.

     *

     * Each value corresponds to the index of the operand in "operands".

     */

neuralnetworks/1.1/types.hal

+14 −6
Original line number Diff line number Diff line
@@ -137,13 +137,19 @@ enum OperationType : @1.0::OperationType { 
     * * 1: A 2-D Tensor of {@link OperandType::TENSOR_INT32}, the paddings

     *      for each spatial dimension of the input tensor. The shape of the

     *      tensor must be {rank(input0), 2}.

     *      padding[i, 0] specifies the number of element to be padded in the

     *      padding[i, 0] specifies the number of elements to be padded in the

     *      front of dimension i.

     *      padding[i, 1] specifies the number of element to be padded after the

     *      padding[i, 1] specifies the number of elements to be padded after the

     *      end of dimension i.

     *

     * Outputs:

     * * 0: A tensor of the same {@link OperandType} as input0.

     * * 0: A tensor of the same {@link OperandType} as input0. The

     *      output tensor has the same rank as input0, and each

     *      dimension of the output tensor has the same size as the

     *      corresponding dimension of the input tensor plus the size

     *      of the padding:

     *          output0.dimension[i] =

     *              padding[i, 0] + input0.dimension[i] + padding[i, 1]

     */

    PAD = 32,
@@ -377,19 +383,21 @@ struct Model { 
    /**

     * All operations included in the model.

     *

     * The operations are sorted into execution order.

     * The operations are sorted into execution order. Every operand

     * with lifetime MODEL_OUTPUT or TEMPORARY_VARIABLE must be

     * written before it is read.

     */

    vec<Operation> operations;



    /**

     * Input indexes of the model.

     * Input indexes of the model. There must be at least one.

     *

     * Each value corresponds to the index of the operand in "operands".

     */

    vec<uint32_t> inputIndexes;



    /**

     * Output indexes of the model.

     * Output indexes of the model. There must be at least one.

     *

     * Each value corresponds to the index of the operand in "operands".

     */