| /hardware/interfaces/neuralnetworks/1.2/ |
| D | types.hal | 232 * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying
259 * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying
281 * * 0: The output 4-D tensor, of shape
303 * * 0 ~ n-1: The list of n input tensors, of shape
314 * tensors. The output shape is [D0, D1, ..., sum(Daxis(i)), ..., Dm].
367 * * 0: A 4-D tensor, of shape [batches, height, width, depth_in],
370 * * 1: A 4-D tensor, of shape
376 * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input
416 * * 0: A 4-D tensor, of shape [batches, height, width, depth_in],
419 * * 1: A 4-D tensor, of shape
[all …]
|
| D | IExecutionCallback.hal | 47 * @param outputShapes A list of shape information of model output operands.
|
| D | types.t | 321 * may not be known is the shape of the input tensors.
435 * Describes the shape information of an output operand after execution.
|
| D | IDevice.hal | 222 * the shape of the tensors, which may only be known at execution time. As
313 * the shape of the tensors, which may only be known at execution time. As
|
| D | IPreparedModel.hal | 135 * @return outputShapes A list of shape information of model output operands.
|
| /hardware/interfaces/neuralnetworks/1.0/ |
| D | types.hal | 147 * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying
170 * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying
188 * * 0: The output 4-D tensor, of shape
208 * * 0 ~ n-1: The list of n input tensors, of shape
218 * tensors. The output shape is [D0, D1, ..., sum(Daxis(i)), ..., Dm].
257 * * 0: A 4-D tensor, of shape [batches, height, width, depth_in],
259 * * 1: A 4-D tensor, of shape
262 * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input
285 * * 0: A 4-D tensor, of shape [batches, height, width, depth_in],
287 * * 1: A 4-D tensor, of shape
[all …]
|
| D | IDevice.hal | 90 * the shape of the tensors, which may only be known at execution time. As
|
| D | types.t | 300 * might not be known is the shape of the input tensors.
|
| /hardware/interfaces/neuralnetworks/1.3/ |
| D | types.hal | 176 * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying
203 * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying
225 * * 0: The output 4-D tensor, of shape
249 * * 0 ~ n-1: The list of n input tensors, of shape
263 * tensors. The output shape is [D0, D1, ..., sum(Daxis(i)), ..., Dm].
330 * * 0: A 4-D tensor, of shape [batches, height, width, depth_in],
333 * * 1: A 4-D tensor, of shape
339 * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input
380 * * 0: A 4-D tensor, of shape [batches, height, width, depth_in],
383 * * 1: A 4-D tensor, of shape
[all …]
|
| D | IExecutionCallback.hal | 53 * @param outputShapes A list of shape information of model output operands.
|
| D | IDevice.hal | 144 * the shape of the tensors, which may only be known at execution time. As
255 * the shape of the tensors, which may only be known at execution time. As
323 * buffer is used as an input, the input shape must be the same as the output shape from the
|
| D | IPreparedModel.hal | 101 * the execution, shape information of model output operands, and
189 * @return outputShapes A list of shape information of model output operands.
|
| D | types.t | 358 * may not be known is the shape of the input tensors.
|
| /hardware/qcom/neuralnetworks/hvxservice/1.0/ |
| D | HexagonUtils.cpp | 279 std::string toString(const ::android::nn::Shape& shape) { in toString() argument
280 return "Shape{.type: " + toString(shape.type) + in toString()
281 ", .dimensions: " + toString(shape.dimensions.data(), shape.dimensions.size()) + in toString()
282 ", .scale: " + std::to_string(shape.scale) + in toString()
283 ", .zeroPoint: " + std::to_string(shape.offset) + "}"; in toString()
|
| D | HexagonModel.h | 86 bool setShape(uint32_t operand, const Shape& shape);
|
| D | HexagonModel.cpp | 117 bool Model::setShape(uint32_t operand, const Shape& shape) { in setShape() argument
121 mOperands[operand].dimensions = shape.dimensions; in setShape()
|
| /hardware/interfaces/neuralnetworks/1.1/ |
| D | types.hal | 33 * dimensions of shape block_shape + [batch], interleaves these blocks back
130 * shape is [1].
149 * for each spatial dimension of the input tensor. The shape of the
173 * a grid of blocks of shape block_shape, and interleaves these blocks with
195 * >= 0. The shape of the tensor must be {M, 2}, where M is the number
210 * Removes dimensions of size 1 from the shape of a tensor.
238 * output shape is [1].
286 * shape is [1].
405 * may not be known is the shape of the input tensors.
|
| D | types.t | 83 * may not be known is the shape of the input tensors.
|
| D | IDevice.hal | 95 * the shape of the tensors, which may only be known at execution time. As
|
| /hardware/interfaces/gnss/1.0/ |
| D | IGnssMeasurementCallback.hal | 569 * - if there is a distorted correlation peak shape, report that multipath
571 * - if there is no distorted correlation peak shape, report
|
| /hardware/interfaces/renderscript/1.0/ |
| D | IContext.hal | 40 * the shape of the window. Any dimensions present in the type must be
|