 |
CMSIS-NN
Version 3.1.0
CMSIS NN Software Library
|
|
CMSIS-NN
Version 3.1.0
CMSIS NN Software Library
|
Macros | |
| #define | USE_INTRINSIC |
Enumerations | |
| enum | arm_nn_activation_type |
| Struct for specifying activation function types. More... | |
Functions | |
| arm_status | arm_convolve_wrapper_s8 (const cmsis_nn_context *ctx, const cmsis_nn_conv_params *conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int32_t *bias_data, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| s8 convolution layer wrapper function with the main purpose to call the optimal kernel available in cmsis-nn to perform the convolution. More... | |
| int32_t | arm_convolve_wrapper_s8_get_buffer_size (const cmsis_nn_conv_params *conv_params, const cmsis_nn_dims *input_dims, const cmsis_nn_dims *filter_dims, const cmsis_nn_dims *output_dims) |
| Get the required buffer size for arm_convolve_wrapper_s8. More... | |
| arm_status | arm_convolve_wrapper_s16 (const cmsis_nn_context *ctx, const cmsis_nn_conv_params *conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q15_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int64_t *bias_data, const cmsis_nn_dims *output_dims, q15_t *output_data) |
| s16 convolution layer wrapper function with the main purpose to call the optimal kernel available in cmsis-nn to perform the convolution. More... | |
| int32_t | arm_convolve_wrapper_s16_get_buffer_size (const cmsis_nn_conv_params *conv_params, const cmsis_nn_dims *input_dims, const cmsis_nn_dims *filter_dims, const cmsis_nn_dims *output_dims) |
| Get the required buffer size for arm_convolve_wrapper_s16. More... | |
| arm_status | arm_convolve_s8 (const cmsis_nn_context *ctx, const cmsis_nn_conv_params *conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int32_t *bias_data, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| Basic s8 convolution function. More... | |
| int32_t | arm_convolve_s8_get_buffer_size (const cmsis_nn_dims *input_dims, const cmsis_nn_dims *filter_dims) |
| Get the required buffer size for s8 convolution function. More... | |
| arm_status | arm_convolve_s16 (const cmsis_nn_context *ctx, const cmsis_nn_conv_params *conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q15_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int64_t *bias_data, const cmsis_nn_dims *output_dims, q15_t *output_data) |
| Basic s16 convolution function. More... | |
| arm_status | arm_convolve_fast_s16 (const cmsis_nn_context *ctx, const cmsis_nn_conv_params *conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q15_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int64_t *bias_data, const cmsis_nn_dims *output_dims, q15_t *output_data) |
| Optimized s16 convolution function. More... | |
| int32_t | arm_convolve_s16_get_buffer_size (const cmsis_nn_dims *input_dims, const cmsis_nn_dims *filter_dims) |
| Get the required buffer size for s16 convolution function. More... | |
| int32_t | arm_convolve_fast_s16_get_buffer_size (const cmsis_nn_dims *input_dims, const cmsis_nn_dims *filter_dims) |
| Get the required buffer size for fast s16 convolution function. More... | |
| arm_status | arm_convolve_HWC_q7_basic (const q7_t *Im_in, const uint16_t dim_im_in, const uint16_t ch_im_in, const q7_t *wt, const uint16_t ch_im_out, const uint16_t dim_kernel, const uint16_t padding, const uint16_t stride, const q7_t *bias, const uint16_t bias_shift, const uint16_t out_shift, q7_t *Im_out, const uint16_t dim_im_out, q15_t *bufferA, q7_t *bufferB) |
| Basic Q7 convolution function. More... | |
| arm_status | arm_convolve_HWC_q7_basic_nonsquare (const q7_t *Im_in, const uint16_t dim_im_in_x, const uint16_t dim_im_in_y, const uint16_t ch_im_in, const q7_t *wt, const uint16_t ch_im_out, const uint16_t dim_kernel_x, const uint16_t dim_kernel_y, const uint16_t padding_x, const uint16_t padding_y, const uint16_t stride_x, const uint16_t stride_y, const q7_t *bias, const uint16_t bias_shift, const uint16_t out_shift, q7_t *Im_out, const uint16_t dim_im_out_x, const uint16_t dim_im_out_y, q15_t *bufferA, q7_t *bufferB) |
| Basic Q7 convolution function (non-square shape) More... | |
| arm_status | arm_convolve_HWC_q15_basic (const q15_t *Im_in, const uint16_t dim_im_in, const uint16_t ch_im_in, const q15_t *wt, const uint16_t ch_im_out, const uint16_t dim_kernel, const uint16_t padding, const uint16_t stride, const q15_t *bias, const uint16_t bias_shift, const uint16_t out_shift, q15_t *Im_out, const uint16_t dim_im_out, q15_t *bufferA, q7_t *bufferB) |
| Basic Q15 convolution function. More... | |
| arm_status | arm_convolve_HWC_q7_fast (const q7_t *Im_in, const uint16_t dim_im_in, const uint16_t ch_im_in, const q7_t *wt, const uint16_t ch_im_out, const uint16_t dim_kernel, const uint16_t padding, const uint16_t stride, const q7_t *bias, const uint16_t bias_shift, const uint16_t out_shift, q7_t *Im_out, const uint16_t dim_im_out, q15_t *bufferA, q7_t *bufferB) |
| Fast Q7 convolution function. More... | |
| arm_status | arm_convolve_HWC_q7_fast_nonsquare (const q7_t *Im_in, const uint16_t dim_im_in_x, const uint16_t dim_im_in_y, const uint16_t ch_im_in, const q7_t *wt, const uint16_t ch_im_out, const uint16_t dim_kernel_x, const uint16_t dim_kernel_y, const uint16_t padding_x, const uint16_t padding_y, const uint16_t stride_x, const uint16_t stride_y, const q7_t *bias, const uint16_t bias_shift, const uint16_t out_shift, q7_t *Im_out, const uint16_t dim_im_out_x, const uint16_t dim_im_out_y, q15_t *bufferA, q7_t *bufferB) |
| Fast Q7 convolution function (non-sqaure shape) More... | |
| arm_status | arm_convolve_1x1_HWC_q7_fast_nonsquare (const q7_t *Im_in, const uint16_t dim_im_in_x, const uint16_t dim_im_in_y, const uint16_t ch_im_in, const q7_t *wt, const uint16_t ch_im_out, const uint16_t dim_kernel_x, const uint16_t dim_kernel_y, const uint16_t padding_x, const uint16_t padding_y, const uint16_t stride_x, const uint16_t stride_y, const q7_t *bias, const uint16_t bias_shift, const uint16_t out_shift, q7_t *Im_out, const uint16_t dim_im_out_x, const uint16_t dim_im_out_y, q15_t *bufferA, q7_t *bufferB) |
| Fast Q7 version of 1x1 convolution (non-sqaure shape) More... | |
| arm_status | arm_convolve_1x1_s8_fast (const cmsis_nn_context *ctx, const cmsis_nn_conv_params *conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int32_t *bias_data, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| Fast s8 version for 1x1 convolution (non-square shape) More... | |
| int32_t | arm_convolve_1x1_s8_fast_get_buffer_size (const cmsis_nn_dims *input_dims) |
| Get the required buffer size for arm_convolve_1x1_s8_fast. More... | |
| arm_status | arm_convolve_1_x_n_s8 (const cmsis_nn_context *ctx, const cmsis_nn_conv_params *conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int32_t *bias_data, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| 1xn convolution More... | |
| int32_t | arm_convolve_1_x_n_s8_get_buffer_size (const cmsis_nn_dims *input_dims, const cmsis_nn_dims *filter_dims) |
| Get the required additional buffer size for 1xn convolution. More... | |
| arm_status | arm_convolve_HWC_q7_RGB (const q7_t *Im_in, const uint16_t dim_im_in, const uint16_t ch_im_in, const q7_t *wt, const uint16_t ch_im_out, const uint16_t dim_kernel, const uint16_t padding, const uint16_t stride, const q7_t *bias, const uint16_t bias_shift, const uint16_t out_shift, q7_t *Im_out, const uint16_t dim_im_out, q15_t *bufferA, q7_t *bufferB) |
| Q7 version of convolution for RGB image. More... | |
| arm_status | arm_convolve_HWC_q15_fast (const q15_t *Im_in, const uint16_t dim_im_in, const uint16_t ch_im_in, const q15_t *wt, const uint16_t ch_im_out, const uint16_t dim_kernel, const uint16_t padding, const uint16_t stride, const q15_t *bias, const uint16_t bias_shift, const uint16_t out_shift, q15_t *Im_out, const uint16_t dim_im_out, q15_t *bufferA, q7_t *bufferB) |
| Fast Q15 convolution function. More... | |
| arm_status | arm_convolve_HWC_q15_fast_nonsquare (const q15_t *Im_in, const uint16_t dim_im_in_x, const uint16_t dim_im_in_y, const uint16_t ch_im_in, const q15_t *wt, const uint16_t ch_im_out, const uint16_t dim_kernel_x, const uint16_t dim_kernel_y, const uint16_t padding_x, const uint16_t padding_y, const uint16_t stride_x, const uint16_t stride_y, const q15_t *bias, const uint16_t bias_shift, const uint16_t out_shift, q15_t *Im_out, const uint16_t dim_im_out_x, const uint16_t dim_im_out_y, q15_t *bufferA, q7_t *bufferB) |
| Fast Q15 convolution function (non-sqaure shape) More... | |
| arm_status | arm_depthwise_separable_conv_HWC_q7 (const q7_t *Im_in, const uint16_t dim_im_in, const uint16_t ch_im_in, const q7_t *wt, const uint16_t ch_im_out, const uint16_t dim_kernel, const uint16_t padding, const uint16_t stride, const q7_t *bias, const uint16_t bias_shift, const uint16_t out_shift, q7_t *Im_out, const uint16_t dim_im_out, q15_t *bufferA, q7_t *bufferB) |
| Q7 depthwise separable convolution function. More... | |
| arm_status | arm_depthwise_separable_conv_HWC_q7_nonsquare (const q7_t *Im_in, const uint16_t dim_im_in_x, const uint16_t dim_im_in_y, const uint16_t ch_im_in, const q7_t *wt, const uint16_t ch_im_out, const uint16_t dim_kernel_x, const uint16_t dim_kernel_y, const uint16_t padding_x, const uint16_t padding_y, const uint16_t stride_x, const uint16_t stride_y, const q7_t *bias, const uint16_t bias_shift, const uint16_t out_shift, q7_t *Im_out, const uint16_t dim_im_out_x, const uint16_t dim_im_out_y, q15_t *bufferA, q7_t *bufferB) |
| Q7 depthwise separable convolution function (non-square shape) More... | |
| arm_status | arm_depthwise_conv_wrapper_s8 (const cmsis_nn_context *ctx, const cmsis_nn_dw_conv_params *dw_conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int32_t *bias_data, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| Wrapper function to pick the right optimized s8 depthwise convolution function. More... | |
| int32_t | arm_depthwise_conv_wrapper_s8_get_buffer_size (const cmsis_nn_dw_conv_params *dw_conv_params, const cmsis_nn_dims *input_dims, const cmsis_nn_dims *filter_dims, const cmsis_nn_dims *output_dims) |
| Get size of additional buffer required by arm_depthwise_conv_wrapper_s8() More... | |
| arm_status | arm_depthwise_conv_s8 (const cmsis_nn_context *ctx, const cmsis_nn_dw_conv_params *dw_conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int32_t *bias_data, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| Basic s8 depthwise convolution function that doesn't have any constraints on the input dimensions. More... | |
| arm_status | arm_depthwise_conv_s16 (const cmsis_nn_context *ctx, const cmsis_nn_dw_conv_params *dw_conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q15_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int64_t *bias_data, const cmsis_nn_dims *output_dims, q15_t *output_data) |
| Basic s16 depthwise convolution function that doesn't have any constraints on the input dimensions. More... | |
| arm_status | arm_depthwise_conv_3x3_s8 (const cmsis_nn_context *ctx, const cmsis_nn_dw_conv_params *dw_conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int32_t *bias_data, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| Optimized s8 depthwise convolution function for 3x3 kernel size with some constraints on the input arguments(documented below). Refer arm_depthwise_conv_s8() for function argument details. More... | |
| arm_status | arm_depthwise_conv_s8_opt (const cmsis_nn_context *ctx, const cmsis_nn_dw_conv_params *dw_conv_params, const cmsis_nn_per_channel_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int32_t *bias_data, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| Optimized s8 depthwise convolution function with constraint that in_channel equals out_channel. Refer arm_depthwise_conv_s8() for function argument details. More... | |
| int32_t | arm_depthwise_conv_s8_opt_get_buffer_size (const cmsis_nn_dims *input_dims, const cmsis_nn_dims *filter_dims) |
| Get the required buffer size for optimized s8 depthwise convolution function with constraint that in_channel equals out_channel. More... | |
| arm_status | arm_fully_connected_q7 (const q7_t *pV, const q7_t *pM, const uint16_t dim_vec, const uint16_t num_of_rows, const uint16_t bias_shift, const uint16_t out_shift, const q7_t *bias, q7_t *pOut, q15_t *vec_buffer) |
| Q7 basic fully-connected layer function. More... | |
| arm_status | arm_fully_connected_s8 (const cmsis_nn_context *ctx, const cmsis_nn_fc_params *fc_params, const cmsis_nn_per_tensor_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int32_t *bias_data, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| Basic s8 Fully Connected function. More... | |
| int32_t | arm_fully_connected_s8_get_buffer_size (const cmsis_nn_dims *filter_dims) |
| Get the required buffer size for S8 basic fully-connected and matrix multiplication layer function for TF Lite. More... | |
| arm_status | arm_fully_connected_s16 (const cmsis_nn_context *ctx, const cmsis_nn_fc_params *fc_params, const cmsis_nn_per_tensor_quant_params *quant_params, const cmsis_nn_dims *input_dims, const q15_t *input_data, const cmsis_nn_dims *filter_dims, const q7_t *filter_data, const cmsis_nn_dims *bias_dims, const int64_t *bias_data, const cmsis_nn_dims *output_dims, q15_t *output_data) |
| Basic s16 Fully Connected function. More... | |
| int32_t | arm_fully_connected_s16_get_buffer_size (const cmsis_nn_dims *filter_dims) |
| Get the required buffer size for S16 basic fully-connected and matrix multiplication layer function for TF Lite. More... | |
| arm_status | arm_fully_connected_q7_opt (const q7_t *pV, const q7_t *pM, const uint16_t dim_vec, const uint16_t num_of_rows, const uint16_t bias_shift, const uint16_t out_shift, const q7_t *bias, q7_t *pOut, q15_t *vec_buffer) |
| Q7 opt fully-connected layer function. More... | |
| arm_status | arm_fully_connected_q15 (const q15_t *pV, const q15_t *pM, const uint16_t dim_vec, const uint16_t num_of_rows, const uint16_t bias_shift, const uint16_t out_shift, const q15_t *bias, q15_t *pOut, q15_t *vec_buffer) |
| Q15 basic fully-connected layer function. More... | |
| arm_status | arm_fully_connected_q15_opt (const q15_t *pV, const q15_t *pM, const uint16_t dim_vec, const uint16_t num_of_rows, const uint16_t bias_shift, const uint16_t out_shift, const q15_t *bias, q15_t *pOut, q15_t *vec_buffer) |
| Q15 opt fully-connected layer function. More... | |
| arm_status | arm_fully_connected_mat_q7_vec_q15 (const q15_t *pV, const q7_t *pM, const uint16_t dim_vec, const uint16_t num_of_rows, const uint16_t bias_shift, const uint16_t out_shift, const q7_t *bias, q15_t *pOut, q15_t *vec_buffer) |
| Mixed Q15-Q7 fully-connected layer function. More... | |
| arm_status | arm_fully_connected_mat_q7_vec_q15_opt (const q15_t *pV, const q7_t *pM, const uint16_t dim_vec, const uint16_t num_of_rows, const uint16_t bias_shift, const uint16_t out_shift, const q7_t *bias, q15_t *pOut, q15_t *vec_buffer) |
| Mixed Q15-Q7 opt fully-connected layer function. More... | |
| q7_t * | arm_nn_mat_mult_kernel_q7_q15 (const q7_t *pA, const q15_t *pInBuffer, const uint16_t ch_im_out, const uint16_t numCol_A, const uint16_t bias_shift, const uint16_t out_shift, const q7_t *bias, q7_t *pOut) |
| Matrix-Multiplication Kernels for Convolution. More... | |
| arm_status | arm_elementwise_add_s8 (const int8_t *input_1_vect, const int8_t *input_2_vect, const int32_t input_1_offset, const int32_t input_1_mult, const int32_t input_1_shift, const int32_t input_2_offset, const int32_t input_2_mult, const int32_t input_2_shift, const int32_t left_shift, int8_t *output, const int32_t out_offset, const int32_t out_mult, const int32_t out_shift, const int32_t out_activation_min, const int32_t out_activation_max, const int32_t block_size) |
| s8 elementwise add of two vectors More... | |
| arm_status | arm_elementwise_add_s16 (const int16_t *input_1_vect, const int16_t *input_2_vect, const int32_t input_1_offset, const int32_t input_1_mult, const int32_t input_1_shift, const int32_t input_2_offset, const int32_t input_2_mult, const int32_t input_2_shift, const int32_t left_shift, int16_t *output, const int32_t out_offset, const int32_t out_mult, const int32_t out_shift, const int32_t out_activation_min, const int32_t out_activation_max, const int32_t block_size) |
| s16 elementwise add of two vectors More... | |
| arm_status | arm_elementwise_mul_s8 (const int8_t *input_1_vect, const int8_t *input_2_vect, const int32_t input_1_offset, const int32_t input_2_offset, int8_t *output, const int32_t out_offset, const int32_t out_mult, const int32_t out_shift, const int32_t out_activation_min, const int32_t out_activation_max, const int32_t block_size) |
| s8 elementwise multiplication More... | |
| arm_status | arm_elementwise_mul_s16 (const int16_t *input_1_vect, const int16_t *input_2_vect, const int32_t input_1_offset, const int32_t input_2_offset, int16_t *output, const int32_t out_offset, const int32_t out_mult, const int32_t out_shift, const int32_t out_activation_min, const int32_t out_activation_max, const int32_t block_size) |
| s16 elementwise multiplication More... | |
| void | arm_relu_q7 (q7_t *data, uint16_t size) |
| Q7 RELU function. More... | |
| void | arm_relu6_s8 (q7_t *data, uint16_t size) |
| s8 ReLU6 function More... | |
| void | arm_relu_q15 (q15_t *data, uint16_t size) |
| Q15 RELU function. More... | |
| void | arm_nn_activations_direct_q7 (q7_t *data, uint16_t size, uint16_t int_width, arm_nn_activation_type type) |
| Q7 neural network activation function using direct table look-up. More... | |
| void | arm_nn_activations_direct_q15 (q15_t *data, uint16_t size, uint16_t int_width, arm_nn_activation_type type) |
| Q15 neural network activation function using direct table look-up. More... | |
| void | arm_maxpool_q7_HWC (q7_t *Im_in, const uint16_t dim_im_in, const uint16_t ch_im_in, const uint16_t dim_kernel, const uint16_t padding, const uint16_t stride, const uint16_t dim_im_out, q7_t *bufferA, q7_t *Im_out) |
| Q7 max pooling function. More... | |
| void | arm_avepool_q7_HWC (q7_t *Im_in, const uint16_t dim_im_in, const uint16_t ch_im_in, const uint16_t dim_kernel, const uint16_t padding, const uint16_t stride, const uint16_t dim_im_out, q7_t *bufferA, q7_t *Im_out) |
| Q7 average pooling function. More... | |
| arm_status | arm_avgpool_s8 (const cmsis_nn_context *ctx, const cmsis_nn_pool_params *pool_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *filter_dims, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| s8 average pooling function. More... | |
| int32_t | arm_avgpool_s8_get_buffer_size (const int dim_dst_width, const int ch_src) |
| Get the required buffer size for S8 average pooling function. More... | |
| arm_status | arm_avgpool_s16 (const cmsis_nn_context *ctx, const cmsis_nn_pool_params *pool_params, const cmsis_nn_dims *input_dims, const int16_t *input_data, const cmsis_nn_dims *filter_dims, const cmsis_nn_dims *output_dims, int16_t *output_data) |
| s16 average pooling function. More... | |
| int32_t | arm_avgpool_s16_get_buffer_size (const int dim_dst_width, const int ch_src) |
| Get the required buffer size for S16 average pooling function. More... | |
| arm_status | arm_max_pool_s8 (const cmsis_nn_context *ctx, const cmsis_nn_pool_params *pool_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *filter_dims, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| s8 max pooling function. More... | |
| arm_status | arm_max_pool_s16 (const cmsis_nn_context *ctx, const cmsis_nn_pool_params *pool_params, const cmsis_nn_dims *input_dims, const int16_t *src, const cmsis_nn_dims *filter_dims, const cmsis_nn_dims *output_dims, int16_t *dst) |
| s16 max pooling function. More... | |
| void | arm_softmax_q7 (const q7_t *vec_in, const uint16_t dim_vec, q7_t *p_out) |
| Q7 softmax function. More... | |
| void | arm_softmax_with_batch_q7 (const q7_t *vec_in, const uint16_t nb_batches, const uint16_t dim_vec, q7_t *p_out) |
| Q7 softmax function with batch parameter. More... | |
| void | arm_softmax_q15 (const q15_t *vec_in, const uint16_t dim_vec, q15_t *p_out) |
| Q15 softmax function. More... | |
| void | arm_softmax_s8 (const int8_t *input, const int32_t num_rows, const int32_t row_size, const int32_t mult, const int32_t shift, const int32_t diff_min, int8_t *output) |
| S8 softmax function. More... | |
| void | arm_softmax_s8_s16 (const int8_t *input, const int32_t num_rows, const int32_t row_size, const int32_t mult, const int32_t shift, const int32_t diff_min, int16_t *output) |
| S8 to s16 softmax function. More... | |
| arm_status | arm_softmax_s16 (const int16_t *input, const int32_t num_rows, const int32_t row_size, const int32_t mult, const int32_t shift, const cmsis_nn_softmax_lut_s16 *softmax_params, int16_t *output) |
| S16 softmax function. More... | |
| void | arm_softmax_u8 (const uint8_t *input, const int32_t num_rows, const int32_t row_size, const int32_t mult, const int32_t shift, const int32_t diff_min, uint8_t *output) |
| U8 softmax function. More... | |
| arm_status | arm_depthwise_conv_u8_basic_ver1 (const uint8_t *input, const uint16_t input_x, const uint16_t input_y, const uint16_t input_ch, const uint8_t *kernel, const uint16_t kernel_x, const uint16_t kernel_y, const int16_t ch_mult, const int16_t pad_x, const int16_t pad_y, const int16_t stride_x, const int16_t stride_y, const int16_t dilation_x, const int16_t dilation_y, const int32_t *bias, const int32_t input_offset, const int32_t filter_offset, const int32_t output_offset, uint8_t *output, const uint16_t output_x, const uint16_t output_y, const int32_t output_activation_min, const int32_t output_activation_max, const int32_t out_shift, const int32_t out_mult) |
| uint8 depthwise convolution function with asymmetric quantization Unless specified otherwise, arguments are mandatory. More... | |
| void | arm_reshape_s8 (const int8_t *input, int8_t *output, const uint32_t total_size) |
| Reshape a s8 vector into another with different shape. More... | |
| void | arm_concatenation_s8_x (const int8_t *input, const uint16_t input_x, const uint16_t input_y, const uint16_t input_z, const uint16_t input_w, int8_t *output, const uint16_t output_x, const uint32_t offset_x) |
| int8/uint8 concatenation function to be used for concatenating N-tensors along the X axis This function should be called for each input tensor to concatenate. The argument offset_x will be used to store the input tensor in the correct position in the output tensor More... | |
| void | arm_concatenation_s8_y (const int8_t *input, const uint16_t input_x, const uint16_t input_y, const uint16_t input_z, const uint16_t input_w, int8_t *output, const uint16_t output_y, const uint32_t offset_y) |
| int8/uint8 concatenation function to be used for concatenating N-tensors along the Y axis This function should be called for each input tensor to concatenate. The argument offset_y will be used to store the input tensor in the correct position in the output tensor More... | |
| void | arm_concatenation_s8_z (const int8_t *input, const uint16_t input_x, const uint16_t input_y, const uint16_t input_z, const uint16_t input_w, int8_t *output, const uint16_t output_z, const uint32_t offset_z) |
| int8/uint8 concatenation function to be used for concatenating N-tensors along the Z axis This function should be called for each input tensor to concatenate. The argument offset_z will be used to store the input tensor in the correct position in the output tensor More... | |
| void | arm_concatenation_s8_w (const int8_t *input, const uint16_t input_x, const uint16_t input_y, const uint16_t input_z, const uint16_t input_w, int8_t *output, const uint32_t offset_w) |
| int8/uint8 concatenation function to be used for concatenating N-tensors along the W axis (Batch size) This function should be called for each input tensor to concatenate. The argument offset_w will be used to store the input tensor in the correct position in the output tensor More... | |
| arm_status | arm_svdf_s8 (const cmsis_nn_context *input_ctx, const cmsis_nn_context *output_ctx, const cmsis_nn_svdf_params *svdf_params, const cmsis_nn_per_tensor_quant_params *input_quant_params, const cmsis_nn_per_tensor_quant_params *output_quant_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *state_dims, q7_t *state_data, const cmsis_nn_dims *weights_feature_dims, const q7_t *weights_feature_data, const cmsis_nn_dims *weights_time_dims, const q7_t *weights_time_data, const cmsis_nn_dims *bias_dims, const q31_t *bias_data, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| s8 SVDF function with 8 bit state tensor and 8 bit time weights More... | |
| arm_status | arm_svdf_state_s16_s8 (const cmsis_nn_context *input_ctx, const cmsis_nn_context *output_ctx, const cmsis_nn_svdf_params *svdf_params, const cmsis_nn_per_tensor_quant_params *input_quant_params, const cmsis_nn_per_tensor_quant_params *output_quant_params, const cmsis_nn_dims *input_dims, const q7_t *input_data, const cmsis_nn_dims *state_dims, q15_t *state_data, const cmsis_nn_dims *weights_feature_dims, const q7_t *weights_feature_data, const cmsis_nn_dims *weights_time_dims, const q15_t *weights_time_data, const cmsis_nn_dims *bias_dims, const q31_t *bias_data, const cmsis_nn_dims *output_dims, q7_t *output_data) |
| s8 SVDF function with 16 bit state tensor and 16 bit time weights More... | |
| #define USE_INTRINSIC |
| arm_status arm_avgpool_s16 | ( | const cmsis_nn_context * | ctx, |
| const cmsis_nn_pool_params * | pool_params, | ||
| const cmsis_nn_dims * | input_dims, | ||
| const int16_t * | input_data, | ||
| const cmsis_nn_dims * | filter_dims, | ||
| const cmsis_nn_dims * | output_dims, | ||
| int16_t * | output_data | ||
| ) |
| [in,out] | ctx | Function context (e.g. temporary buffer). Check the function definition file to see if an additional buffer is required. Optional function {API}_get_buffer_size() provides the buffer size if an additional buffer is required. |
| [in] | pool_params | Pooling parameters |
| [in] | input_dims | Input (activation) tensor dimensions. Format: [H, W, C_IN] Argument 'N' is not used. |
| [in] | input_data | Input (activation) data pointer. Data type: int16 |
| [in] | filter_dims | Filter tensor dimensions. Format: [H, W] Argument N and C are not used. |
| [in] | output_dims | Output tensor dimensions. Format: [H, W, C_OUT] Argument N is not used. C_OUT equals C_IN. |
| [in,out] | output_data | Output data pointer. Data type: int16 |
ARM_MATH_SUCCESS - Successful operation| q7_t* arm_nn_mat_mult_kernel_q7_q15 | ( | const q7_t * | pA, |
| const q15_t * | pInBuffer, | ||
| const uint16_t | ch_im_out, | ||
| const uint16_t | numCol_A, | ||
| const uint16_t | bias_shift, | ||
| const uint16_t | out_shift, | ||
| const q7_t * | bias, | ||
| q7_t * | pOut | ||
| ) |
These functions are used within convolution layer functions for matrix multiplication.
The implementation is similar to CMSIS-DSP arm_mat_mult functions with one Q7 and one Q15 operands. The Q15 operand is the im2col output which is always with 2 columns. Matrix-multiplication function for convolution
| [in] | pA | pointer to operand A |
| [in] | pInBuffer | pointer to operand B, always conssists of 2 vectors |
| [in] | ch_im_out | numRow of A |
| [in] | numCol_A | numCol of A |
| [in] | bias_shift | amount of left-shift for bias |
| [in] | out_shift | amount of right-shift for output |
| [in] | bias | the bias |
| [in,out] | pOut | pointer to output |
Matrix-Multiplication Kernels for Convolution.
Refer to header file for details.
References arm_nn_read_q15x2_ia(), and NN_ROUND.
Referenced by arm_convolve_HWC_q7_basic(), arm_convolve_HWC_q7_basic_nonsquare(), and arm_convolve_HWC_q7_RGB().