1 /** 2 * Copyright 2020-2022 Huawei Technologies Co., Ltd 3 * 4 * Licensed under the Apache License, Version 2.0 (the "License"); 5 * you may not use this file except in compliance with the License. 6 * You may obtain a copy of the License at 7 * 8 * http://www.apache.org/licenses/LICENSE-2.0 9 * 10 * Unless required by applicable law or agreed to in writing, software 11 * distributed under the License is distributed on an "AS IS" BASIS, 12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 13 * See the License for the specific language governing permissions and 14 * limitations under the License. 15 */ 16 17 #ifndef MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_IR_DATASETOPS_SOURCE_SAMPLERS_DISTRIBUTED_SAMPLER_IR_H_ 18 #define MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_IR_DATASETOPS_SOURCE_SAMPLERS_DISTRIBUTED_SAMPLER_IR_H_ 19 20 #include <memory> 21 #include <string> 22 #include <nlohmann/json.hpp> 23 24 #include "minddata/dataset/engine/ir/datasetops/source/samplers/samplers_ir.h" 25 #include "include/api/status.h" 26 #ifndef ENABLE_ANDROID 27 #include "minddata/mindrecord/include/shard_operator.h" 28 #endif 29 30 namespace mindspore { 31 namespace dataset { 32 // Internal Sampler class forward declaration 33 class SamplerRT; 34 35 class DistributedSamplerObj : public SamplerObj { 36 public: 37 DistributedSamplerObj(int64_t num_shards, int64_t shard_id, bool shuffle, int64_t num_samples, uint32_t seed, 38 int64_t offset, bool even_dist); 39 40 ~DistributedSamplerObj() override; 41 42 Status SamplerBuild(std::shared_ptr<SamplerRT> *sampler) override; 43 44 std::shared_ptr<SamplerObj> SamplerCopy() override; 45 46 #ifndef ENABLE_ANDROID 47 std::shared_ptr<mindrecord::ShardOperator> BuildForMindDataset() override; 48 #endif 49 50 /// \brief Get the arguments of node 51 /// \param[out] out_json JSON string of all attributes 52 /// \return Status of the function 53 Status to_json(nlohmann::json *const out_json) override; 54 55 #ifndef ENABLE_ANDROID 56 /// \brief Function for read sampler from JSON object 57 /// \param[in] json_obj JSON object to be read 58 /// \param[in] num_samples number of sample in the sampler 59 /// \param[out] sampler Sampler constructed from parameters in JSON object 60 /// \return Status of the function 61 static Status from_json(nlohmann::json json_obj, int64_t num_samples, std::shared_ptr<SamplerObj> *sampler); 62 #endif 63 64 Status ValidateParams() override; 65 66 /// \brief Function to get the shard id of sampler 67 /// \return The shard id of sampler 68 int64_t ShardId() override; 69 70 private: 71 int64_t num_shards_; 72 int64_t shard_id_; 73 bool shuffle_; 74 int64_t num_samples_; 75 uint32_t seed_; 76 int64_t offset_; 77 bool even_dist_; 78 }; 79 } // namespace dataset 80 } // namespace mindspore 81 #endif // MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_IR_DATASETOPS_SOURCE_SAMPLERS_DISTRIBUTED_SAMPLER_IR_H_ 82