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 17 #ifndef MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_TREE_ADAPTER_LITE_H_ 18 #define MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_TREE_ADAPTER_LITE_H_ 19 20 #include <memory> 21 #include <string> 22 #include <unordered_map> 23 #include <utility> 24 #include <vector> 25 26 #include "minddata/dataset/engine/ir/datasetops/dataset_node.h" 27 28 namespace mindspore { 29 namespace dataset { 30 31 class TensorRow; 32 class DatasetNode; 33 34 class TreeAdapterLite { 35 public: 36 // this flag is used to indicate the purpose of the creation of this tree adapter (type of the tree_consumer). 37 // Currently there are 3 types of consumer, Iterator, Getter and TDT/Vocab/Save ... 38 // To avoid premature optimization, the last type (TDT/Vocab/Save) is regarded as Iterator for now. 39 enum UsageFlag { kDeIterator = 0, kDeGetter = 1, kDeReset = 2 }; 40 41 explicit TreeAdapterLite(UsageFlag usage = kDeGetter); 42 43 ~TreeAdapterLite() = default; 44 45 Status BuildTree(std::shared_ptr<DatasetNode> root_ir); 46 47 // Get rows equal to num_rows 48 Status GetNextRow(TensorRow *const row); 49 GetColumnNameMap()50 std::unordered_map<std::string, int32_t> GetColumnNameMap() const { return tree_->root()->column_name_id_map(); } 51 52 // unique_ptr overloads operator bool(), will return false if it doesn't manage an object GetRoot()53 std::weak_ptr<DatasetOp> GetRoot() const { return tree_ ? tree_->root() : nullptr; } 54 55 // This function performs syntax checking, semantics checking, and then call BuildTree 56 Status Compile(const std::shared_ptr<DatasetNode> &input_ir, int32_t num_epochs = -1); 57 58 protected: 59 // Run the mandatory pass checking the syntax and semantics of the IR tree 60 Status PrePass(std::shared_ptr<DatasetNode> ir); 61 62 // Run the mandatory pass augmenting the IR tree 63 Status PostPass(std::shared_ptr<DatasetNode> ir) const; 64 65 // Return Offload Json 66 nlohmann::json GetOffloadJson(); 67 68 std::shared_ptr<DatasetNode> input_ir_; 69 std::shared_ptr<DatasetNode> root_ir_; 70 71 private: 72 // This RECURSIVE function walks the (optimized) IR tree in DFS to build its corresponding Execution tree. 73 Status BuildExecutionTreeRecur(std::shared_ptr<DatasetNode> ir, std::shared_ptr<DatasetOp> *op); 74 75 std::shared_ptr<DatasetOp> root_; // current connector capacity of root op, used for profiling 76 std::unique_ptr<ExecutionTree> tree_; 77 UsageFlag usage_; // usage of this tree adapter (type of consumer) 78 nlohmann::json offload_json_; 79 }; 80 81 } // namespace dataset 82 } // namespace mindspore 83 84 #endif // MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_TREE_ADAPTER_LITE_H_ 85