| /* |
| * Copyright (C) 2019 The Android Open Source Project |
| * |
| * Licensed under the Apache License, Version 2.0 (the "License"); |
| * you may not use this file except in compliance with the License. |
| * You may obtain a copy of the License at |
| * |
| * http://www.apache.org/licenses/LICENSE-2.0 |
| * |
| * Unless required by applicable law or agreed to in writing, software |
| * distributed under the License is distributed on an "AS IS" BASIS, |
| * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| * See the License for the specific language governing permissions and |
| * limitations under the License. |
| */ |
| |
| #include "fuzzing/operation_signatures/OperationSignatureUtils.h" |
| |
| namespace android { |
| namespace nn { |
| namespace fuzzing_test { |
| |
| static void softmaxConstructor(Type dataType, uint32_t rank, RandomOperation* op) { |
| sameDimensionOpConstructor(dataType, rank, op); |
| // Generate value for "axis" parameter. |
| if (op->inputs.size() > 2) { |
| op->inputs[2]->setScalarValue<int32_t>(getUniform<int32_t>(-rank, rank - 1)); |
| } |
| } |
| |
| DEFINE_OPERATION_SIGNATURE(SOFTMAX_V1_0){ |
| .opType = ANEURALNETWORKS_SOFTMAX, |
| .supportedDataTypes = {Type::TENSOR_FLOAT32, Type::TENSOR_QUANT8_ASYMM}, |
| .supportedRanks = {2, 4}, |
| .version = HalVersion::V1_0, |
| .inputs = {INPUT_DEFAULT, PARAMETER_FLOAT_RANGE(0.1, 10.0)}, |
| .outputs = {OUTPUT_QUANT(1.f / 256, 0)}, |
| .constructor = softmaxConstructor}; |
| |
| DEFINE_OPERATION_SIGNATURE(SOFTMAX_V1_2){ |
| .opType = ANEURALNETWORKS_SOFTMAX, |
| .supportedDataTypes = {Type::TENSOR_FLOAT32, Type::TENSOR_QUANT8_ASYMM}, |
| .supportedRanks = {1, 3}, |
| .version = HalVersion::V1_2, |
| .inputs = {INPUT_DEFAULT, PARAMETER_FLOAT_RANGE(0.1, 10.0)}, |
| .outputs = {OUTPUT_QUANT(1.f / 256, 0)}, |
| .constructor = softmaxConstructor}; |
| |
| DEFINE_OPERATION_SIGNATURE(SOFTMAX_float16_V1_2){ |
| .opType = ANEURALNETWORKS_SOFTMAX, |
| .supportedDataTypes = {Type::TENSOR_FLOAT16}, |
| .supportedRanks = {1, 2, 3, 4}, |
| .version = HalVersion::V1_2, |
| .inputs = {INPUT_DEFAULT, PARAMETER_FLOAT_RANGE(0.1, 10.0)}, |
| .outputs = {OUTPUT_QUANT(1.f / 256, 0)}, |
| .constructor = softmaxConstructor}; |
| |
| DEFINE_OPERATION_SIGNATURE(SOFTMAX_axis_V1_2){ |
| .opType = ANEURALNETWORKS_SOFTMAX, |
| .supportedDataTypes = {Type::TENSOR_FLOAT32, Type::TENSOR_QUANT8_ASYMM, |
| Type::TENSOR_FLOAT16}, |
| .supportedRanks = {1, 2, 3, 4}, |
| .version = HalVersion::V1_2, |
| .inputs = {INPUT_DEFAULT, PARAMETER_FLOAT_RANGE(0.1, 10.0), PARAMETER_NONE(Type::INT32)}, |
| .outputs = {OUTPUT_QUANT(1.f / 256, 0)}, |
| .constructor = softmaxConstructor}; |
| |
| static void l2normConstructor(Type dataType, uint32_t rank, RandomOperation* op) { |
| sameDimensionOpConstructor(dataType, rank, op); |
| // Generate value for "axis" parameter. |
| if (op->inputs.size() > 1) { |
| op->inputs[1]->setScalarValue<int32_t>(getUniform<int32_t>(-rank, rank - 1)); |
| } |
| // L2_NORMALIZATION may produce NaN output values with all zero inputs. We should not connect |
| // the output tensor to the input of another operation. |
| op->outputs[0]->doNotConnect = true; |
| } |
| |
| DEFINE_OPERATION_SIGNATURE(L2_NORMALIZATION_V1_0){.opType = ANEURALNETWORKS_L2_NORMALIZATION, |
| .supportedDataTypes = {Type::TENSOR_FLOAT32}, |
| .supportedRanks = {4}, |
| .version = HalVersion::V1_0, |
| .inputs = {INPUT_DEFAULT}, |
| .outputs = {OUTPUT_DEFAULT}, |
| .constructor = l2normConstructor}; |
| |
| DEFINE_OPERATION_SIGNATURE(L2_NORMALIZATION_V1_2){.opType = ANEURALNETWORKS_L2_NORMALIZATION, |
| .supportedDataTypes = {Type::TENSOR_FLOAT32}, |
| .supportedRanks = {1, 2, 3}, |
| .version = HalVersion::V1_2, |
| .inputs = {INPUT_DEFAULT}, |
| .outputs = {OUTPUT_DEFAULT}, |
| .constructor = l2normConstructor}; |
| |
| DEFINE_OPERATION_SIGNATURE(L2_NORMALIZATION_type_V1_2){ |
| .opType = ANEURALNETWORKS_L2_NORMALIZATION, |
| .supportedDataTypes = {Type::TENSOR_FLOAT16, Type::TENSOR_QUANT8_ASYMM}, |
| .supportedRanks = {1, 2, 3, 4}, |
| .version = HalVersion::V1_2, |
| .inputs = {INPUT_DEFAULT}, |
| .outputs = {OUTPUT_QUANT(1.f / 128, 128)}, |
| .constructor = l2normConstructor}; |
| |
| DEFINE_OPERATION_SIGNATURE(L2_NORMALIZATION_axis_V1_2){ |
| .opType = ANEURALNETWORKS_L2_NORMALIZATION, |
| .supportedDataTypes = {Type::TENSOR_FLOAT32, Type::TENSOR_FLOAT16, |
| Type::TENSOR_QUANT8_ASYMM}, |
| .supportedRanks = {1, 2, 3, 4}, |
| .version = HalVersion::V1_2, |
| .inputs = {INPUT_DEFAULT, PARAMETER_NONE(Type::INT32)}, |
| .outputs = {OUTPUT_QUANT(1.f / 128, 128)}, |
| .constructor = l2normConstructor}; |
| |
| static void localResponseNormConstructor(Type dataType, uint32_t rank, RandomOperation* op) { |
| sameDimensionOpConstructor(dataType, rank, op); |
| // Generate value for "axis" parameter. |
| if (op->inputs.size() > 5) { |
| op->inputs[5]->setScalarValue<int32_t>(getUniform<int32_t>(-rank, rank - 1)); |
| } |
| } |
| |
| DEFINE_OPERATION_SIGNATURE(LOCAL_RESPONSE_NORMALIZATION_V1_0){ |
| .opType = ANEURALNETWORKS_LOCAL_RESPONSE_NORMALIZATION, |
| .supportedDataTypes = {Type::TENSOR_FLOAT32}, |
| .supportedRanks = {4}, |
| .version = HalVersion::V1_0, |
| .inputs = |
| { |
| INPUT_DEFAULT, |
| PARAMETER_RANGE(Type::INT32, 1, 10), |
| PARAMETER_FLOAT_RANGE(0.0, 10.0), |
| PARAMETER_FLOAT_RANGE(0.1, 10.0), |
| PARAMETER_FLOAT_RANGE(0.1, 1.0), |
| }, |
| .outputs = {OUTPUT_DEFAULT}, |
| .constructor = localResponseNormConstructor}; |
| |
| DEFINE_OPERATION_SIGNATURE(LOCAL_RESPONSE_NORMALIZATION_V1_2){ |
| .opType = ANEURALNETWORKS_LOCAL_RESPONSE_NORMALIZATION, |
| .supportedDataTypes = {Type::TENSOR_FLOAT32}, |
| .supportedRanks = {1, 2, 3}, |
| .version = HalVersion::V1_2, |
| .inputs = |
| { |
| INPUT_DEFAULT, |
| PARAMETER_RANGE(Type::INT32, 1, 10), |
| PARAMETER_FLOAT_RANGE(0.0, 10.0), |
| PARAMETER_FLOAT_RANGE(0.1, 10.0), |
| PARAMETER_FLOAT_RANGE(0.1, 1.0), |
| }, |
| .outputs = {OUTPUT_DEFAULT}, |
| .constructor = localResponseNormConstructor}; |
| |
| DEFINE_OPERATION_SIGNATURE(LOCAL_RESPONSE_NORMALIZATION_float16_V1_2){ |
| .opType = ANEURALNETWORKS_LOCAL_RESPONSE_NORMALIZATION, |
| .supportedDataTypes = {Type::TENSOR_FLOAT16}, |
| .supportedRanks = {1, 2, 3, 4}, |
| .version = HalVersion::V1_2, |
| .inputs = |
| { |
| INPUT_DEFAULT, |
| PARAMETER_RANGE(Type::INT32, 1, 10), |
| PARAMETER_FLOAT_RANGE(0.0, 10.0), |
| PARAMETER_FLOAT_RANGE(0.1, 10.0), |
| PARAMETER_FLOAT_RANGE(0.1, 1.0), |
| }, |
| .outputs = {OUTPUT_DEFAULT}, |
| .constructor = localResponseNormConstructor}; |
| |
| DEFINE_OPERATION_SIGNATURE(LOCAL_RESPONSE_NORMALIZATION_axis_V1_2){ |
| .opType = ANEURALNETWORKS_LOCAL_RESPONSE_NORMALIZATION, |
| .supportedDataTypes = {Type::TENSOR_FLOAT32, Type::TENSOR_FLOAT16}, |
| .supportedRanks = {1, 2, 3, 4}, |
| .version = HalVersion::V1_2, |
| .inputs = |
| { |
| INPUT_DEFAULT, |
| PARAMETER_RANGE(Type::INT32, 1, 10), |
| PARAMETER_FLOAT_RANGE(0.0, 10.0), |
| PARAMETER_FLOAT_RANGE(0.1, 10.0), |
| PARAMETER_FLOAT_RANGE(0.1, 1.0), |
| PARAMETER_NONE(Type::INT32), |
| }, |
| .outputs = {OUTPUT_DEFAULT}, |
| .constructor = localResponseNormConstructor}; |
| |
| DEFINE_OPERATION_SIGNATURE(INSTANCE_NORMALIZATION_V1_2){ |
| .opType = ANEURALNETWORKS_INSTANCE_NORMALIZATION, |
| .supportedDataTypes = {Type::TENSOR_FLOAT32, Type::TENSOR_FLOAT16}, |
| .supportedRanks = {4}, |
| .version = HalVersion::V1_2, |
| .inputs = |
| { |
| INPUT_DEFAULT, |
| PARAMETER_FLOAT_RANGE(0.1, 10.0), |
| PARAMETER_FLOAT_RANGE(-10.0, 10.0), |
| PARAMETER_FLOAT_RANGE(0.01, 1.0), |
| PARAMETER_CHOICE(Type::BOOL, true, false), |
| }, |
| .outputs = {OUTPUT_DEFAULT}, |
| .constructor = sameShapeOpConstructor}; |
| |
| DEFINE_OPERATION_SIGNATURE(LOG_SOFTMAX_TENSOR_FLOAT32_V1_2){ |
| .opType = ANEURALNETWORKS_LOG_SOFTMAX, |
| .supportedDataTypes = {Type::TENSOR_FLOAT32, Type::TENSOR_FLOAT16}, |
| .supportedRanks = {1, 2, 3, 4}, |
| .version = HalVersion::V1_2, |
| .inputs = {INPUT_DEFAULT, PARAMETER_FLOAT_RANGE(0.1, 10.0), PARAMETER_NONE(Type::INT32)}, |
| .outputs = {OUTPUT_DEFAULT}, |
| .constructor = softmaxConstructor}; |
| |
| } // namespace fuzzing_test |
| } // namespace nn |
| } // namespace android |
