1 /** 2 * Copyright 2021 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 #ifndef MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_DATASETOPS_SOURCE_SAMPLER_MINDRECORD_SAMPLER_H_ 17 #define MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_DATASETOPS_SOURCE_SAMPLER_MINDRECORD_SAMPLER_H_ 18 19 #include <limits> 20 #include <memory> 21 #include <vector> 22 23 #include "minddata/dataset/engine/datasetops/source/sampler/sampler.h" 24 #include "minddata/mindrecord/include/shard_reader.h" 25 26 namespace mindspore { 27 namespace dataset { 28 class MindRecordSamplerRT : public SamplerRT { 29 public: 30 // Constructor 31 // @param shard_reader - shard_reader 32 // @param int64_t samples_per_tensor - Num of Sampler Ids to fetch via 1 GetNextSample call 33 MindRecordSamplerRT(mindrecord::ShardReader *shard_reader, 34 int64_t samples_per_tensor = std::numeric_limits<int64_t>::max()); 35 36 // Destructor. 37 ~MindRecordSamplerRT() = default; 38 39 // Op calls this to get next set of sampleIds 40 // @param out - Tensor of sample ids to be returned to caller 41 // @return Status The status code returned 42 Status GetNextSample(TensorRow *out) override; 43 44 // meant to be called by base class or python 45 Status InitSampler() override; 46 47 /// \brief Reset for next epoch. 48 /// \param[in] failover_reset A boolean to show whether we are resetting the pipeline 49 /// \return Status The status code returned 50 Status ResetSampler(const bool failover_reset = false) override; 51 52 void SamplerPrint(std::ostream &out, bool show_all) const override; 53 54 /// \brief Get the arguments of node 55 /// \param[out] out_json JSON string of all attributes 56 /// \return Status of the function 57 Status to_json(nlohmann::json *out_json) override; 58 59 private: 60 mindrecord::ShardReader *shard_reader_; // back pointer to the shard reader 61 const std::vector<int64_t> *sample_ids_; // read-only back pointer into mind record sampler ids 62 int64_t next_id_; 63 }; 64 } // namespace dataset 65 } // namespace mindspore 66 #endif // MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_DATASETOPS_SOURCE_SAMPLER_MINDRECORD_SAMPLER_H_ 67