diff options
Diffstat (limited to 'src/cpu/operators/CpuFullyConnected.cpp')
| -rw-r--r-- | src/cpu/operators/CpuFullyConnected.cpp | 50 |
1 files changed, 39 insertions, 11 deletions
diff --git a/src/cpu/operators/CpuFullyConnected.cpp b/src/cpu/operators/CpuFullyConnected.cpp index 1e1598a8ee..af630154cf 100644 --- a/src/cpu/operators/CpuFullyConnected.cpp +++ b/src/cpu/operators/CpuFullyConnected.cpp @@ -166,7 +166,8 @@ CpuFullyConnected::CpuFullyConnected() _is_prepared(false), _enable_fast_math(false), _fixed_format(false), - _weight_format(arm_compute::WeightFormat::UNSPECIFIED) + _weight_format(arm_compute::WeightFormat::UNSPECIFIED), + _dynamic_weights(false) { } @@ -189,7 +190,7 @@ void CpuFullyConnected::configure_mm(const ITensorInfo *src, const ITensorInfo * const Status status = get_gemmlowp_output_stage_info(&src_info, &weights_info, dst, act, gemmlowp_output_stage_info); ARM_COMPUTE_ERROR_ON(status.error_code() != ErrorCode::OK); - GEMMInfo gemm_info; + GEMMInfo gemm_info(false, false, !_dynamic_weights /* Reshape weights only for the first run */); gemm_info.set_gemmlowp_output_stage(gemmlowp_output_stage_info); gemm_info.set_activation_info(act); gemm_info.set_fast_math(_enable_fast_math); @@ -199,7 +200,7 @@ void CpuFullyConnected::configure_mm(const ITensorInfo *src, const ITensorInfo * else { // Configure matrix multiply kernel - GEMMInfo gemm_info(false, false, true /* Reshape weights only for the first run */); + GEMMInfo gemm_info(false, false, !_dynamic_weights /* Reshape weights only for the first run */); gemm_info.set_activation_info(act); gemm_info.set_fast_math(_enable_fast_math); gemm_info.set_fixed_format(_fixed_format); @@ -256,6 +257,7 @@ void CpuFullyConnected::configure(const ITensorInfo *src, const ITensorInfo *wei _enable_fast_math = fc_info.enable_fast_math; _fixed_format = weights_info.weight_format() != WeightFormat::UNSPECIFIED; _weight_format = weights_info.weight_format(); + _dynamic_weights = !weights->are_values_constant() && _needs_weights_reshape; // With the Fully Connected layer we can have 4 different cases: // 1) Convolution layer -> Fully Connected layer without batches @@ -329,15 +331,32 @@ void CpuFullyConnected::configure(const ITensorInfo *src, const ITensorInfo *wei { // Release permuted weights at the end of prepare as they are further transposed by the assembly dispatch // Do not release them if biases are dynamic and data type is quantized, since the weights tensor will be used for biases offset calculation - _aux_mem[TransposedWeights] = MemoryInfo(offset_int_vec(TransposedWeights), (_is_quantized_asymmetric && biases - && !(biases->are_values_constant())) ? MemoryLifetime::Persistent : MemoryLifetime::Prepare, - _reshaped_weights.total_size()); - _aux_mem[ConvertedWeights] = MemoryInfo(offset_int_vec(ConvertedWeights), MemoryLifetime::Prepare, _converted_weights.total_size()); + // Keep all the auxiliary tensors in case of dynamic weights as they are recalculated every time. + _aux_mem[TransposedWeights] = MemoryInfo( + offset_int_vec(TransposedWeights), + _dynamic_weights ? MemoryLifetime::Temporary : + (_is_quantized_asymmetric && biases && !(biases->are_values_constant())) ? MemoryLifetime::Persistent : + MemoryLifetime::Prepare, + _reshaped_weights.total_size()); + + _aux_mem[ConvertedWeights] = MemoryInfo( + offset_int_vec(ConvertedWeights), + _dynamic_weights ? MemoryLifetime::Temporary : MemoryLifetime::Prepare, + _converted_weights.total_size()); } else { - _aux_mem[TransposedWeights] = MemoryInfo(offset_int_vec(TransposedWeights), _needs_weights_conversion ? MemoryLifetime::Prepare : MemoryLifetime::Persistent, _reshaped_weights.total_size()); - _aux_mem[ConvertedWeights] = MemoryInfo(offset_int_vec(ConvertedWeights), MemoryLifetime::Persistent, _converted_weights.total_size()); + _aux_mem[TransposedWeights] = MemoryInfo( + offset_int_vec(TransposedWeights), + _dynamic_weights ? MemoryLifetime::Temporary : + _needs_weights_conversion ? MemoryLifetime::Prepare : + MemoryLifetime::Persistent, + _reshaped_weights.total_size()); + + _aux_mem[ConvertedWeights] = MemoryInfo( + offset_int_vec(ConvertedWeights), + _dynamic_weights ? MemoryLifetime::Temporary : MemoryLifetime::Persistent, + _converted_weights.total_size()); } _aux_mem[FlattenedSrc] = MemoryInfo(offset_int_vec(FlattenedSrc), MemoryLifetime::Temporary, _flattened_src.total_size()); } @@ -375,7 +394,6 @@ Status CpuFullyConnected::validate(const ITensorInfo *src, const ITensorInfo *we ARM_COMPUTE_RETURN_ERROR_ON(weights->num_dimensions() > 2); ARM_COMPUTE_RETURN_ERROR_ON(fc_info.activation_info.enabled() && is_data_type_quantized(src->data_type()) && fc_info.activation_info.activation() != ActivationLayerInfo::ActivationFunction::RELU && fc_info.activation_info.activation() != ActivationLayerInfo::ActivationFunction::BOUNDED_RELU && fc_info.activation_info.activation() != ActivationLayerInfo::ActivationFunction::LU_BOUNDED_RELU); - ARM_COMPUTE_RETURN_ERROR_ON(!weights->are_values_constant() && (!fc_info.are_weights_reshaped || fc_info.transpose_weights)); bool weights_reshaped = fc_info.transpose_weights ? fc_info.are_weights_reshaped : true; bool is_fc_after_conv = true; @@ -459,6 +477,11 @@ void CpuFullyConnected::run(ITensorPack &tensors) { prepare(tensors); +#ifdef ARM_COMPUTE_ASSERTS_ENABLED + ++_asrt_run_count; + ARM_COMPUTE_ERROR_ON(_dynamic_weights && _asrt_prepare_count != _asrt_run_count); +#endif // ARM_COMPUTE_ASSERTS_ENABLED + auto src = tensors.get_const_tensor(ACL_SRC_0); CpuAuxTensorHandler flattened_src(offset_int_vec(FlattenedSrc), _flattened_src, tensors, false); @@ -491,8 +514,13 @@ void CpuFullyConnected::run(ITensorPack &tensors) void CpuFullyConnected::prepare(ITensorPack &tensors) { - if(!_is_prepared) + if(!_is_prepared || _dynamic_weights) { +#ifdef ARM_COMPUTE_ASSERTS_ENABLED + ++_asrt_prepare_count; + ARM_COMPUTE_ERROR_ON(!_dynamic_weights && _asrt_prepare_count > 1); +#endif // ARM_COMPUTE_ASSERTS_ENABLED + auto weights = tensors.get_const_tensor(ACL_SRC_1); CpuAuxTensorHandler reshaped_weights(offset_int_vec(TransposedWeights), _reshaped_weights, tensors, false); |