1 /** 2 * Copyright 2020 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_TEXT_KERNELS_BASIC_TOKENIZER_OP_H_ 17 #define MINDSPORE_CCSRC_MINDDATA_DATASET_TEXT_KERNELS_BASIC_TOKENIZER_OP_H_ 18 #include <memory> 19 #include <string> 20 #include <unordered_set> 21 22 #include "minddata/dataset/core/tensor.h" 23 #include "minddata/dataset/kernels/tensor_op.h" 24 #include "minddata/dataset/text/kernels/case_fold_op.h" 25 #include "minddata/dataset/text/kernels/normalize_utf8_op.h" 26 #include "minddata/dataset/text/kernels/regex_replace_op.h" 27 #include "minddata/dataset/text/kernels/regex_tokenizer_op.h" 28 #include "minddata/dataset/text/kernels/tokenizer_op.h" 29 #include "minddata/dataset/util/status.h" 30 31 namespace mindspore { 32 namespace dataset { 33 34 class BasicTokenizerOp : public TokenizerOp { 35 public: 36 static const bool kDefLowerCase; 37 static const bool kDefKeepWhitespace; 38 static const NormalizeForm kDefNormalizationForm; 39 static const bool kDefPreserveUnusedToken; 40 41 explicit BasicTokenizerOp(const bool &lower_case = kDefLowerCase, const bool &keep_whitespace = kDefKeepWhitespace, 42 const NormalizeForm &normalization_form = kDefNormalizationForm, 43 const bool &preserve_unused_token = kDefPreserveUnusedToken, 44 const bool &with_offsets = kDefWithOffsets); 45 46 ~BasicTokenizerOp() override = default; 47 48 Status Compute(const TensorRow &input, TensorRow *output) override; 49 50 protected: 51 Status CaseFoldWithoutUnusedWords(const std::string_view &text, const std::unordered_set<std::string> &unused_words, 52 std::string *output); 53 Status CaseFoldWithoutUnusedWords(const std::shared_ptr<Tensor> &input, std::shared_ptr<Tensor> *output); 54 Name()55 std::string Name() const override { return kBasicTokenizerOp; } 56 57 private: 58 static const char kCommonPattern[]; 59 static const char kUnusedPattern[]; 60 static const std::unordered_set<std::string> kUnusedWords; 61 bool lower_case_; 62 bool keep_whitespace_; 63 NormalizeForm normalization_form_; 64 bool preserve_unused_token_; 65 std::unique_ptr<CaseFoldOp> case_fold_; 66 std::unique_ptr<NormalizeUTF8Op> nfd_normalize_; 67 std::unique_ptr<NormalizeUTF8Op> common_normalize_; 68 std::unique_ptr<RegexReplaceOp> replace_accent_chars_; 69 std::unique_ptr<RegexReplaceOp> replace_control_chars_; 70 std::unique_ptr<RegexTokenizerOp> regex_tokenizer_; 71 }; 72 } // namespace dataset 73 } // namespace mindspore 74 #endif // MINDSPORE_CCSRC_MINDDATA_DATASET_TEXT_KERNELS_BASIC_TOKENIZER_OP_H_ 75