- Parameters
-
[in] | input_ctx | Temporary scratch buffer |
[in] | output_ctx | Temporary output scratch buffer |
[in] | svdf_params | SVDF Parameters Range of svdf_params->input_offset : [-128, 127] Range of svdf_params->output_offset : [-128, 127] |
[in] | input_quant_params | Input quantization parameters |
[in] | output_quant_params | Output quantization parameters |
[in] | input_dims | Input tensor dimensions |
[in] | input_data | Pointer to input tensor |
[in] | state_dims | State tensor dimensions |
[in] | state_data | Pointer to state tensor |
[in] | weights_feature_dims | Weights (feature) tensor dimensions |
[in] | weights_feature_data | Pointer to the weights (feature) tensor |
[in] | weights_time_dims | Weights (time) tensor dimensions |
[in] | weights_time_data | Pointer to the weights (time) tensor |
[in] | bias_dims | Bias tensor dimensions |
[in] | bias_data | Pointer to bias tensor |
[in] | output_dims | Output tensor dimensions |
[out] | output_data | Pointer to the output tensor |
- Returns
- The function returns
ARM_MATH_SUCCESS
- Supported framework: TensorFlow Lite micro
- q7 is used as data type eventhough it is s8 data. It is done so to be consistent with existing APIs.
References arm_nn_read_q15x2_ia(), arm_nn_requantize(), arm_nn_vec_mat_mult_t_svdf_s8(), cmsis_nn_context::buf, CLAMP, cmsis_nn_dims::h, cmsis_nn_svdf_params::input_activation, cmsis_nn_svdf_params::input_offset, cmsis_nn_activation::max, cmsis_nn_activation::min, cmsis_nn_per_tensor_quant_params::multiplier, cmsis_nn_dims::n, cmsis_nn_svdf_params::output_activation, cmsis_nn_svdf_params::output_offset, cmsis_nn_svdf_params::rank, and cmsis_nn_per_tensor_quant_params::shift.