Home
last modified time | relevance | path

Searched refs:depth_out (Results 1 – 5 of 5) sorted by relevance

/hardware/qcom/neuralnetworks/hvxservice/1.0/
DHexagonController.cpp148 uint32_t* width_out, uint32_t* depth_out, uint8_t* data_out, in execute() argument
151 batches_out, height_out, width_out, depth_out, data_out, data_out_max, in execute()
DHexagonController.h82 uint32_t* depth_out, uint8_t* data_out, uint32_t data_out_max,
/hardware/qcom/neuralnetworks/hvxservice/1.0/hexagon_nn_controller/
Dhexagon_nn_controller.h108 unsigned int* depth_out, unsigned char* data_out,
/hardware/interfaces/neuralnetworks/1.0/
Dtypes.hal253 * [depth_out, filter_height, filter_width, depth_in], specifying the
255 * * 2: A 1-D tensor, of shape [depth_out], specifying the bias.
281 * [depth_out, filter_height, filter_width, depth_in], specifying the
283 * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input
302 * [batches, out_height, out_width, depth_out]. For output tensor of
314 * filter tensor of shape [1, filter_height, filter_width, depth_out]
315 * containing depth_out convolutional filters of depth 1, DEPTHWISE_CONV
320 * The output has depth_out = depth_in * depth_multiplier channels.
343 * * 1: A 4-D tensor, of shape [1, filter_height, filter_width, depth_out],
345 * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input
[all …]
/hardware/interfaces/neuralnetworks/1.2/
Dtypes.hal379 * [depth_out, filter_height, filter_width, depth_in], specifying the
383 * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input
427 * [depth_out, filter_height, filter_width, depth_in], specifying the
431 * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input
467 * [batches, out_height, out_width, depth_out]. Before API level 29,
480 * filter tensor of shape [1, filter_height, filter_width, depth_out]
481 * containing depth_out convolutional filters of depth 1, DEPTHWISE_CONV
486 * The output has depth_out = depth_in * depth_multiplier channels.
527 * * 1: A 4-D tensor, of shape [1, filter_height, filter_width, depth_out],
531 * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input
[all …]