enhance: add example for zhwiki and CLUERNER2020 to mindrecord

example/cv_to_mindrecord/ImageNet_Similar_Perf/README.md

@@ -35,11 +35,20 @@ This example provides an efficient way to generate MindRecord. Users only need t
     n02110185 3 mouse
     n02096294 4 orange
     ```
+
 2. Edit run_imagenet.sh and modify the parameters
-3. Run the bash script  
+    ```
+    --mindrecord_file: output MindRecord file.
+    --mindrecord_partitions: the partitions for MindRecord.
+    --label_file: ImageNet label map file.
+    --image_dir: ImageNet dir which contain sub dir.
+    ```
+
+3. Run the bash script
     ```bash  
     bash run_imagenet.sh
     ```  
+
 4. Performance result
 
     |  Training Data |  General API | Current Example |  Env  |
@@ -47,29 +56,30 @@ This example provides an efficient way to generate MindRecord. Users only need t
     |ImageNet(140G)|  2h40m |  50m  |  CPU: Intel Xeon Gold 6130 x 64, Memory: 256G, Storage: HDD |
 
 ## How to use the example for other dataset
+
 ### Create work space
 
 Assume the dataset name is 'xyz'
 * Create work space from template
     ```shell
-    cd ${your_mindspore_home}/example/convert_to_mindrecord
+    cd ${your_mindspore_home}/example/cv_to_mindrecord/ImageNet_Similar_Perf
     cp -r template xyz
     ```
 
 ### Implement data generator
 
-Edit dictionary data generator  
+Edit dictionary data generator.
 * Edit file 
     ```shell
-    cd ${your_mindspore_home}/example/convert_to_mindrecord
+    cd ${your_mindspore_home}/example/cv_to_mindrecord/ImageNet_Similar_Perf
     vi xyz/mr_api.py
     ```
 
-Two API, 'mindrecord_task_number' and 'mindrecord_dict_data', must be implemented
+Two API, 'mindrecord_task_number' and 'mindrecord_dict_data', must be implemented.
 - 'mindrecord_task_number()' returns number of tasks. Return 1 if data row is generated serially. Return N if generator can be split into N parallel-run tasks.
 - 'mindrecord_dict_data(task_id)' yields dictionary data row by row. 'task_id' is 0..N-1, if N is return value of mindrecord_task_number()
 
-Tricky for parallel run
+Tricky for parallel run.
 - For ImageNet, one directory can be a task.
 - For TFRecord with multiple files, each file can be a task.
 - For TFRecord with 1 file only, it could also be split into N tasks. Task_id=K means: data row is picked only if (count % N == K) 
@@ -78,7 +88,7 @@ Tricky for parallel run
 
 * run python script 
     ```shell
-    cd ${your_mindspore_home}/example/convert_to_mindrecord
+    cd ${your_mindspore_home}/example/cv_to_mindrecord/ImageNet_Similar_Perf
     python writer.py --mindrecord_script xyz [...]
     ```
     > You can put this command in script **run_xyz.sh** for easy execution

example/nlp_to_mindrecord/CLUERNER2020/README.md

@@ -0,0 +1,82 @@
+# Guideline to Convert Training Data CLUERNER2020 to MindRecord For Bert Fine Tuning
+
+<!-- TOC -->
+
+- [What does the example do](#what-does-the-example-do)
+- [How to use the example to process CLUERNER2020](#how-to-use-the-example-to-process-cluerner2020)
+    - [Download CLUERNER2020 and unzip](#download-cluerner2020-and-unzip)
+    - [Generate MindRecord](#generate-mindrecord)
+    - [Create MindDataset By MindRecord](#create-minddataset-by-mindrecord)
+
+
+<!-- /TOC -->
+
+## What does the example do
+
+This example is based on [CLUERNER2020](https://www.cluebenchmarks.com/introduce.html) training data, generating MindRecord file, and finally used for Bert Fine Tuning progress.
+
+1.  run.sh: generate MindRecord entry script
+    - data_processor_seq.py: the script from [CLUEbenchmark/CLUENER2020](https://github.com/CLUEbenchmark/CLUENER2020/tree/master/tf_version), we just change the part of the generated tfrecord to MindRecord.
+    - label2id.json: the file from [CLUEbenchmark/CLUENER2020](https://github.com/CLUEbenchmark/CLUENER2020/tree/master/tf_version).
+    - tokenization.py: the script from [CLUEbenchmark/CLUENER2020](https://github.com/CLUEbenchmark/CLUENER2020/tree/master/tf_version).
+    - vocab.txt: the file from [CLUEbenchmark/CLUENER2020](https://github.com/CLUEbenchmark/CLUENER2020/tree/master/tf_version).
+2.  run_read.py: create MindDataset by MindRecord entry script.
+    - create_dataset.py: use MindDataset to read MindRecord to generate dataset.
+3. data: the output directory for MindRecord.
+4. cluener_public: the CLUENER2020 training data.
+
+## How to use the example to process CLUERNER2020
+
+Download CLUERNER2020, convert it to MindRecord, use MindDataset to read MindRecord.
+
+### Download CLUERNER2020 and unzip
+
+1. Download the training data zip.
+    > [CLUERNER2020 dataset download address](https://www.cluebenchmarks.com/introduce.html) **-> 任务介绍 -> CLUENER 细粒度命名实体识别 -> cluener下载链接**
+
+2. Unzip the training data to dir example/nlp_to_mindrecord/CLUERNER2020/cluener_public.
+    ```
+    unzip -d {your-mindspore}/example/nlp_to_mindrecord/CLUERNER2020/cluener_public cluener_public.zip
+    ```
+
+### Generate MindRecord
+
+1. Run the run.sh script.
+    ```bash
+    bash run.sh
+    ```
+
+2. Output like this:
+    ```
+    ...
+    [INFO] ME(17603:139620983514944,MainProcess):2020-04-28-16:56:12.498.235 [mindspore/mindrecord/filewriter.py:313] The list of mindrecord files created are: ['data/train.mindrecord'], and the list of index files are: ['data/train.mindrecord.db']
+    ...
+    [INFO] ME(17603,python):2020-04-28-16:56:13.400.175 [mindspore/ccsrc/mindrecord/io/shard_writer.cc:667] WriteRawData] Write 1 records successfully.
+    [INFO] ME(17603,python):2020-04-28-16:56:13.400.863 [mindspore/ccsrc/mindrecord/io/shard_writer.cc:667] WriteRawData] Write 1 records successfully.
+    [INFO] ME(17603,python):2020-04-28-16:56:13.401.534 [mindspore/ccsrc/mindrecord/io/shard_writer.cc:667] WriteRawData] Write 1 records successfully.
+    [INFO] ME(17603,python):2020-04-28-16:56:13.402.179 [mindspore/ccsrc/mindrecord/io/shard_writer.cc:667] WriteRawData] Write 1 records successfully.
+    [INFO] ME(17603,python):2020-04-28-16:56:13.402.702 [mindspore/ccsrc/mindrecord/io/shard_writer.cc:667] WriteRawData] Write 1 records successfully.
+    ...
+    [INFO] ME(17603:139620983514944,MainProcess):2020-04-28-16:56:13.431.208 [mindspore/mindrecord/filewriter.py:313] The list of mindrecord files created are: ['data/dev.mindrecord'], and the list of index files are: ['data/dev.mindrecord.db']
+    ```
+
+### Create MindDataset By MindRecord
+
+1. Run the run_read.sh script.
+    ```bash
+    bash run_read.sh
+    ```
+
+2. Output like this:
+    ```
+    ...
+    example 1340: input_ids: [ 101 3173 1290 4852 7676 3949  122 3299  123  126 3189 4510 8020 6381 5442 7357 2590 3636 8021 7676 3949 4294 1166 6121 3124 1277 6121 3124 7270 2135 3295 5789 3326 123  126 3189 1355 6134 1093 1325 3173 2399 6590 6791 8024  102    0    0    0    0    0    0    0    0    0    0   0    0    0    0    0    0    0    0]
+    example 1340: input_mask: [1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1  1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
+    example 1340: segment_ids: [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
+    example 1340: label_ids: [ 0 18 19 20  2  4  0  0  0  0  0  0  0 34 36 26 27 28  0 34 35 35 35 35 35 35 35 35 35 36 26 27 28  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0]
+    example 1341: input_ids: [ 101 1728  711 4293 3868 1168 2190 2150 3791  934 3633 3428 4638 6237 7025 8024 3297 1400 5310 3362 6206 5023 5401 1744 3297 7770 3791 7368  976 1139 1104 2137  511 102    0    0    0    0    0    0    0    0   0    0    0    0    0    0    0    0    0    0    0    0    0    0   0    0    0    0    0    0    0    0]
+    example 1341: input_mask: [1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0  0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
+    example 1341: segment_ids: [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
+   example 1341: label_ids: [ 0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0 18 19 19 19 19 20  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0]
+    ...
+    ```

example/nlp_to_mindrecord/CLUERNER2020/create_dataset.py

@@ -0,0 +1,36 @@
+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# 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.
+# ============================================================================
+"""create MindDataset by MindRecord"""
+import mindspore.dataset as ds
+
+def create_dataset(data_file):
+    """create MindDataset"""
+    num_readers = 4
+    data_set = ds.MindDataset(dataset_file=data_file, num_parallel_workers=num_readers, shuffle=True)
+    index = 0
+    for item in data_set.create_dict_iterator():
+        # print("example {}: {}".format(index, item))
+        print("example {}: input_ids: {}".format(index, item['input_ids']))
+        print("example {}: input_mask: {}".format(index, item['input_mask']))
+        print("example {}: segment_ids: {}".format(index, item['segment_ids']))
+        print("example {}: label_ids: {}".format(index, item['label_ids']))
+        index += 1
+        if index % 1000 == 0:
+            print("read rows: {}".format(index))
+    print("total rows: {}".format(index))
+
+if __name__ == '__main__':
+    create_dataset('data/train.mindrecord')
+    create_dataset('data/dev.mindrecord')

example/nlp_to_mindrecord/CLUERNER2020/data/README.md

@@ -0,0 +1 @@
+## output dir

example/nlp_to_mindrecord/CLUERNER2020/data_processor_seq.py

@@ -0,0 +1,162 @@
+#!/usr/bin/python
+# coding:utf8
+"""
+@author: Cong Yu
+@time: 2019-12-07 17:03
+"""
+import json
+import tokenization
+import collections
+
+import numpy as np
+from mindspore.mindrecord import FileWriter
+
+# pylint: skip-file
+
+def _truncate_seq_pair(tokens_a, tokens_b, max_length):
+    """Truncates a sequence pair in place to the maximum length."""
+
+    # This is a simple heuristic which will always truncate the longer sequence
+    # one token at a time. This makes more sense than truncating an equal percent
+    # of tokens from each, since if one sequence is very short then each token
+    # that's truncated likely contains more information than a longer sequence.
+    while True:
+        total_length = len(tokens_a) + len(tokens_b)
+        if total_length <= max_length:
+            break
+        if len(tokens_a) > len(tokens_b):
+            tokens_a.pop()
+        else:
+            tokens_b.pop()
+
+
+def process_one_example(tokenizer, label2id, text, label, max_seq_len=128):
+    # textlist = text.split(' ')
+    # labellist = label.split(' ')
+    textlist = list(text)
+    labellist = list(label)
+    tokens = []
+    labels = []
+    for i, word in enumerate(textlist):
+        token = tokenizer.tokenize(word)
+        tokens.extend(token)
+        label_1 = labellist[i]
+        for m in range(len(token)):
+            if m == 0:
+                labels.append(label_1)
+            else:
+                print("some unknown token...")
+                labels.append(labels[0])
+    # tokens = tokenizer.tokenize(example.text)  -2 的原因是因为序列需要加一个句首和句尾标志
+    if len(tokens) >= max_seq_len - 1:
+        tokens = tokens[0:(max_seq_len - 2)]
+        labels = labels[0:(max_seq_len - 2)]
+    ntokens = []
+    segment_ids = []
+    label_ids = []
+    ntokens.append("[CLS]")  # 句子开始设置CLS 标志
+    segment_ids.append(0)
+    # [CLS] [SEP] 可以为 他们构建标签，或者 统一到某个标签，反正他们是不变的，基本不参加训练 即：x-l 永远不变
+    label_ids.append(0)  # label2id["[CLS]"]
+    for i, token in enumerate(tokens):
+        ntokens.append(token)
+        segment_ids.append(0)
+        label_ids.append(label2id[labels[i]])
+    ntokens.append("[SEP]")
+    segment_ids.append(0)
+    # append("O") or append("[SEP]") not sure!
+    label_ids.append(0)  # label2id["[SEP]"]
+    input_ids = tokenizer.convert_tokens_to_ids(ntokens)
+    input_mask = [1] * len(input_ids)
+    while len(input_ids) < max_seq_len:
+        input_ids.append(0)
+        input_mask.append(0)
+        segment_ids.append(0)
+        label_ids.append(0)
+        ntokens.append("**NULL**")
+    assert len(input_ids) == max_seq_len
+    assert len(input_mask) == max_seq_len
+    assert len(segment_ids) == max_seq_len
+    assert len(label_ids) == max_seq_len
+
+    feature = (input_ids, input_mask, segment_ids, label_ids)
+    return feature
+
+
+def prepare_mindrecord_data(tokenizer, max_seq_len, label2id, path, out_path):
+    """
+        生成训练数据， *.mindrecord, 单标签分类模型, 随机打乱数据
+    """
+    writer = FileWriter(out_path)
+
+    data_schema = {"input_ids": {"type": "int64", "shape": [-1]},
+                   "input_mask": {"type": "int64", "shape": [-1]},
+                   "segment_ids": {"type": "int64", "shape": [-1]},
+                   "label_ids": {"type": "int64", "shape": [-1]}}
+    writer.add_schema(data_schema, "CLUENER2020 schema")
+
+    example_count = 0
+
+    for line in open(path):
+        if not line.strip():
+            continue
+        _ = json.loads(line.strip())
+        len_ = len(_["text"])
+        labels = ["O"] * len_
+        for k, v in _["label"].items():
+            for kk, vv in v.items():
+                for vvv in vv:
+                    span = vvv
+                    s = span[0]
+                    e = span[1] + 1
+                    # print(s, e)
+                    if e - s == 1:
+                        labels[s] = "S_" + k
+                    else:
+                        labels[s] = "B_" + k
+                        for i in range(s + 1, e - 1):
+                            labels[i] = "M_" + k
+                        labels[e - 1] = "E_" + k
+            # print()
+        # feature = process_one_example(tokenizer, label2id, row[column_name_x1], row[column_name_y],
+        #                               max_seq_len=max_seq_len)
+        feature = process_one_example(tokenizer, label2id, list(_["text"]), labels,
+                                      max_seq_len=max_seq_len)
+
+        features = collections.OrderedDict()
+        # 序列标注任务
+        features["input_ids"] = np.asarray(feature[0])
+        features["input_mask"] = np.asarray(feature[1])
+        features["segment_ids"] = np.asarray(feature[2])
+        features["label_ids"] = np.asarray(feature[3])
+        if example_count < 5:
+            print("*** Example ***")
+            print(_["text"])
+            print(_["label"])
+            print("input_ids: %s" % " ".join([str(x) for x in feature[0]]))
+            print("input_mask: %s" % " ".join([str(x) for x in feature[1]]))
+            print("segment_ids: %s" % " ".join([str(x) for x in feature[2]]))
+            print("label: %s " % " ".join([str(x) for x in feature[3]]))
+
+        writer.write_raw_data([features])
+        example_count += 1
+
+        # if example_count == 20:
+        #     break
+        if example_count % 3000 == 0:
+            print(example_count)
+    print("total example:", example_count)
+    writer.commit()
+
+
+if __name__ == "__main__":
+    vocab_file = "./vocab.txt"
+    tokenizer = tokenization.FullTokenizer(vocab_file=vocab_file)
+    label2id = json.loads(open("label2id.json").read())
+
+    max_seq_len = 64
+
+    prepare_mindrecord_data(tokenizer, max_seq_len, label2id, path="cluener_public/train.json",
+                           out_path="data/train.mindrecord")
+    prepare_mindrecord_data(tokenizer, max_seq_len, label2id, path="cluener_public/dev.json",
+                           out_path="data/dev.mindrecord")

example/nlp_to_mindrecord/CLUERNER2020/label2id.json

@@ -0,0 +1,43 @@
+{
+  "O": 0,
+  "S_address": 1,
+  "B_address": 2,
+  "M_address": 3,
+  "E_address": 4,
+  "S_book": 5,
+  "B_book": 6,
+  "M_book": 7,
+  "E_book": 8,
+  "S_company": 9,
+  "B_company": 10,
+  "M_company": 11,
+  "E_company": 12,
+  "S_game": 13,
+  "B_game": 14,
+  "M_game": 15,
+  "E_game": 16,
+  "S_government": 17,
+  "B_government": 18,
+  "M_government": 19,
+  "E_government": 20,
+  "S_movie": 21,
+  "B_movie": 22,
+  "M_movie": 23,
+  "E_movie": 24,
+  "S_name": 25,
+  "B_name": 26,
+  "M_name": 27,
+  "E_name": 28,
+  "S_organization": 29,
+  "B_organization": 30,
+  "M_organization": 31,
+  "E_organization": 32,
+  "S_position": 33,
+  "B_position": 34,
+  "M_position": 35,
+  "E_position": 36,
+  "S_scene": 37,
+  "B_scene": 38,
+  "M_scene": 39,
+  "E_scene": 40
+}

example/nlp_to_mindrecord/CLUERNER2020/run.sh

@@ -0,0 +1,20 @@
+#!/bin/bash
+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# 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.
+# ============================================================================
+
+rm data/train.mindrecord*
+rm data/dev.mindrecord*
+
+python data_processor_seq.py

example/nlp_to_mindrecord/CLUERNER2020/run_read.sh

@@ -0,0 +1,17 @@
+#!/bin/bash
+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# 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.
+# ============================================================================
+
+python create_dataset.py

example/nlp_to_mindrecord/CLUERNER2020/tokenization.py

@@ -0,0 +1,388 @@
+"""Tokenization classes."""
+
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import collections
+import re
+import unicodedata
+import six
+
+# pylint: skip-file
+
+def validate_case_matches_checkpoint(do_lower_case, init_checkpoint):
+    """Checks whether the casing config is consistent with the checkpoint name."""
+
+    # The casing has to be passed in by the user and there is no explicit check
+    # as to whether it matches the checkpoint. The casing information probably
+    # should have been stored in the bert_config.json file, but it's not, so
+    # we have to heuristically detect it to validate.
+
+    if not init_checkpoint:
+        return
+
+    m = re.match("^.*?([A-Za-z0-9_-]+)/bert_model.ckpt", init_checkpoint)
+    if m is None:
+        return
+
+    model_name = m.group(1)
+
+    lower_models = [
+        "uncased_L-24_H-1024_A-16", "uncased_L-12_H-768_A-12",
+        "multilingual_L-12_H-768_A-12", "chinese_L-12_H-768_A-12"
+    ]
+
+    cased_models = [
+        "cased_L-12_H-768_A-12", "cased_L-24_H-1024_A-16",
+        "multi_cased_L-12_H-768_A-12"
+    ]
+
+    is_bad_config = False
+    if model_name in lower_models and not do_lower_case:
+        is_bad_config = True
+        actual_flag = "False"
+        case_name = "lowercased"
+        opposite_flag = "True"
+
+    if model_name in cased_models and do_lower_case:
+        is_bad_config = True
+        actual_flag = "True"
+        case_name = "cased"
+        opposite_flag = "False"
+
+    if is_bad_config:
+        raise ValueError(
+            "You passed in `--do_lower_case=%s` with `--init_checkpoint=%s`. "
+            "However, `%s` seems to be a %s model, so you "
+            "should pass in `--do_lower_case=%s` so that the fine-tuning matches "
+            "how the model was pre-training. If this error is wrong, please "
+            "just comment out this check." % (actual_flag, init_checkpoint,
+                                              model_name, case_name, opposite_flag))
+
+
+def convert_to_unicode(text):
+    """Converts `text` to Unicode (if it's not already), assuming utf-8 input."""
+    if six.PY3:
+        if isinstance(text, str):
+            return text
+        elif isinstance(text, bytes):
+            return text.decode("utf-8", "ignore")
+        else:
+            raise ValueError("Unsupported string type: %s" % (type(text)))
+    elif six.PY2:
+        if isinstance(text, str):
+            return text.decode("utf-8", "ignore")
+        elif isinstance(text, unicode):
+            return text
+        else:
+            raise ValueError("Unsupported string type: %s" % (type(text)))
+    else:
+        raise ValueError("Not running on Python2 or Python 3?")
+
+
+def printable_text(text):
+    """Returns text encoded in a way suitable for print or `tf.logging`."""
+
+    # These functions want `str` for both Python2 and Python3, but in one case
+    # it's a Unicode string and in the other it's a byte string.
+    if six.PY3:
+        if isinstance(text, str):
+            return text
+        elif isinstance(text, bytes):
+            return text.decode("utf-8", "ignore")
+        else:
+            raise ValueError("Unsupported string type: %s" % (type(text)))
+    elif six.PY2:
+        if isinstance(text, str):
+            return text
+        elif isinstance(text, unicode):
+            return text.encode("utf-8")
+        else:
+            raise ValueError("Unsupported string type: %s" % (type(text)))
+    else:
+        raise ValueError("Not running on Python2 or Python 3?")
+
+
+def load_vocab(vocab_file):
+    """Loads a vocabulary file into a dictionary."""
+    vocab = collections.OrderedDict()
+    index = 0
+    with open(vocab_file, "r") as reader:
+        while True:
+            token = convert_to_unicode(reader.readline())
+            if not token:
+                break
+            token = token.strip()
+            vocab[token] = index
+            index += 1
+    return vocab
+
+
+def convert_by_vocab(vocab, items):
+    """Converts a sequence of [tokens|ids] using the vocab."""
+    output = []
+    for item in items:
+        if item in vocab:
+            output.append(vocab[item])
+        else:
+            output.append(vocab['[UNK]'])
+    return output
+
+
+def convert_tokens_to_ids(vocab, tokens):
+    return convert_by_vocab(vocab, tokens)
+
+
+def convert_ids_to_tokens(inv_vocab, ids):
+    return convert_by_vocab(inv_vocab, ids)
+
+
+def whitespace_tokenize(text):
+    """Runs basic whitespace cleaning and splitting on a piece of text."""
+    text = text.strip()
+    if not text:
+        return []
+    tokens = text.split()
+    return tokens
+
+
+class FullTokenizer(object):
+    """Runs end-to-end tokenziation."""
+
+    def __init__(self, vocab_file, do_lower_case=True):
+        self.vocab = load_vocab(vocab_file)
+        self.inv_vocab = {v: k for k, v in self.vocab.items()}
+        self.basic_tokenizer = BasicTokenizer(do_lower_case=do_lower_case)
+        self.wordpiece_tokenizer = WordpieceTokenizer(vocab=self.vocab)
+
+    def tokenize(self, text):
+        split_tokens = []
+        for token in self.basic_tokenizer.tokenize(text):
+            for sub_token in self.wordpiece_tokenizer.tokenize(token):
+                split_tokens.append(sub_token)
+
+        return split_tokens
+
+    def convert_tokens_to_ids(self, tokens):
+        return convert_by_vocab(self.vocab, tokens)
+
+    def convert_ids_to_tokens(self, ids):
+        return convert_by_vocab(self.inv_vocab, ids)
+
+
+class BasicTokenizer(object):
+    """Runs basic tokenization (punctuation splitting, lower casing, etc.)."""
+
+    def __init__(self, do_lower_case=True):
+        """Constructs a BasicTokenizer.
+
+        Args:
+          do_lower_case: Whether to lower case the input.
+        """
+        self.do_lower_case = do_lower_case
+
+    def tokenize(self, text):
+        """Tokenizes a piece of text."""
+        text = convert_to_unicode(text)
+        text = self._clean_text(text)
+
+        # This was added on November 1st, 2018 for the multilingual and Chinese
+        # models. This is also applied to the English models now, but it doesn't
+        # matter since the English models were not trained on any Chinese data
+        # and generally don't have any Chinese data in them (there are Chinese
+        # characters in the vocabulary because Wikipedia does have some Chinese
+        # words in the English Wikipedia.).
+        text = self._tokenize_chinese_chars(text)
+
+        orig_tokens = whitespace_tokenize(text)
+        split_tokens = []
+        for token in orig_tokens:
+            if self.do_lower_case:
+                token = token.lower()
+                token = self._run_strip_accents(token)
+            split_tokens.extend(self._run_split_on_punc(token))
+
+        output_tokens = whitespace_tokenize(" ".join(split_tokens))
+        return output_tokens
+
+    def _run_strip_accents(self, text):
+        """Strips accents from a piece of text."""
+        text = unicodedata.normalize("NFD", text)
+        output = []
+        for char in text:
+            cat = unicodedata.category(char)
+            if cat == "Mn":
+                continue
+            output.append(char)
+        return "".join(output)
+
+    def _run_split_on_punc(self, text):
+        """Splits punctuation on a piece of text."""
+        chars = list(text)
+        i = 0
+        start_new_word = True
+        output = []
+        while i < len(chars):
+            char = chars[i]
+            if _is_punctuation(char):
+                output.append([char])
+                start_new_word = True
+            else:
+                if start_new_word:
+                    output.append([])
+                start_new_word = False
+                output[-1].append(char)
+            i += 1
+
+        return ["".join(x) for x in output]
+
+    def _tokenize_chinese_chars(self, text):
+        """Adds whitespace around any CJK character."""
+        output = []
+        for char in text:
+            cp = ord(char)
+            if self._is_chinese_char(cp):
+                output.append(" ")
+                output.append(char)
+                output.append(" ")
+            else:
+                output.append(char)
+        return "".join(output)
+
+    def _is_chinese_char(self, cp):
+        """Checks whether CP is the codepoint of a CJK character."""
+        # This defines a "chinese character" as anything in the CJK Unicode block:
+        #   https://en.wikipedia.org/wiki/CJK_Unified_Ideographs_(Unicode_block)
+        #
+        # Note that the CJK Unicode block is NOT all Japanese and Korean characters,
+        # despite its name. The modern Korean Hangul alphabet is a different block,
+        # as is Japanese Hiragana and Katakana. Those alphabets are used to write
+        # space-separated words, so they are not treated specially and handled
+        # like the all of the other languages.
+        if ((cp >= 0x4E00 and cp <= 0x9FFF) or  #
+                (cp >= 0x3400 and cp <= 0x4DBF) or  #
+                (cp >= 0x20000 and cp <= 0x2A6DF) or  #
+                (cp >= 0x2A700 and cp <= 0x2B73F) or  #
+                (cp >= 0x2B740 and cp <= 0x2B81F) or  #
+                (cp >= 0x2B820 and cp <= 0x2CEAF) or
+                (cp >= 0xF900 and cp <= 0xFAFF) or  #
+                (cp >= 0x2F800 and cp <= 0x2FA1F)):  #
+            return True
+
+        return False
+
+    def _clean_text(self, text):
+        """Performs invalid character removal and whitespace cleanup on text."""
+        output = []
+        for char in text:
+            cp = ord(char)
+            if cp == 0 or cp == 0xfffd or _is_control(char):
+                continue
+            if _is_whitespace(char):
+                output.append(" ")
+            else:
+                output.append(char)
+        return "".join(output)
+
+
+class WordpieceTokenizer(object):
+    """Runs WordPiece tokenziation."""
+
+    def __init__(self, vocab, unk_token="[UNK]", max_input_chars_per_word=200):
+        self.vocab = vocab
+        self.unk_token = unk_token
+        self.max_input_chars_per_word = max_input_chars_per_word
+
+    def tokenize(self, text):
+        """Tokenizes a piece of text into its word pieces.
+
+        This uses a greedy longest-match-first algorithm to perform tokenization
+        using the given vocabulary.
+
+        For example:
+          input = "unaffable"
+          output = ["un", "##aff", "##able"]
+
+        Args:
+          text: A single token or whitespace separated tokens. This should have
+            already been passed through `BasicTokenizer.
+
+        Returns:
+          A list of wordpiece tokens.
+        """
+
+        text = convert_to_unicode(text)
+
+        output_tokens = []
+        for token in whitespace_tokenize(text):
+            chars = list(token)
+            if len(chars) > self.max_input_chars_per_word:
+                output_tokens.append(self.unk_token)
+                continue
+
+            is_bad = False
+            start = 0
+            sub_tokens = []
+            while start < len(chars):
+                end = len(chars)
+                cur_substr = None
+                while start < end:
+                    substr = "".join(chars[start:end])
+                    if start > 0:
+                        substr = "##" + substr
+                    if substr in self.vocab:
+                        cur_substr = substr
+                        break
+                    end -= 1
+                if cur_substr is None:
+                    is_bad = True
+                    break
+                sub_tokens.append(cur_substr)
+                start = end
+
+            if is_bad:
+                output_tokens.append(self.unk_token)
+            else:
+                output_tokens.extend(sub_tokens)
+        return output_tokens
+
+
+def _is_whitespace(char):
+    """Checks whether `chars` is a whitespace character."""
+    # \t, \n, and \r are technically contorl characters but we treat them
+    # as whitespace since they are generally considered as such.
+    if char == " " or char == "\t" or char == "\n" or char == "\r":
+        return True
+    cat = unicodedata.category(char)
+    if cat == "Zs":
+        return True
+    return False
+
+
+def _is_control(char):
+    """Checks whether `chars` is a control character."""
+    # These are technically control characters but we count them as whitespace
+    # characters.
+    if char == "\t" or char == "\n" or char == "\r":
+        return False
+    cat = unicodedata.category(char)
+    if cat.startswith("C"):
+        return True
+    return False
+
+
+def _is_punctuation(char):
+    """Checks whether `chars` is a punctuation character."""
+    cp = ord(char)
+    # We treat all non-letter/number ASCII as punctuation.
+    # Characters such as "^", "$", and "`" are not in the Unicode
+    # Punctuation class but we treat them as punctuation anyways, for
+    # consistency.
+    if ((cp >= 33 and cp <= 47) or (cp >= 58 and cp <= 64) or
+            (cp >= 91 and cp <= 96) or (cp >= 123 and cp <= 126)):
+        return True
+    cat = unicodedata.category(char)
+    if cat.startswith("P"):
+        return True
+    return False

example/nlp_to_mindrecord/zhwiki/README.md

@@ -0,0 +1,107 @@
+# Guideline to Convert Training Data zhwiki to MindRecord For Bert Pre Training
+
+<!-- TOC -->
+
+- [What does the example do](#what-does-the-example-do)
+- [Run simple test](#run-simple-test)
+- [How to use the example to process zhwiki](#how-to-use-the-example-to-process-zhwiki)
+    - [Download zhwiki training data](#download-zhwiki-training-data)
+    - [Extract the zhwiki](#extract-the-zhwiki)
+    - [Generate MindRecord](#generate-mindrecord)
+    - [Create MindDataset By MindRecord](#create-minddataset-by-mindrecord)
+
+
+<!-- /TOC -->
+
+## What does the example do
+
+This example is based on [zhwiki](https://dumps.wikimedia.org/zhwiki) training data, generating MindRecord file, and finally used for Bert network training.
+
+1.  run.sh: generate MindRecord entry script.
+    - create_pretraining_data.py: the script from [google-research/bert](https://github.com/google-research/bert), we just change the part of the generated tfrecord to MindRecord.
+    - tokenization.py: the script from [google-research/bert](https://github.com/google-research/bert).
+    - vocab.txt: the file from [huawei-noah/Pretrained-Language-Model](https://github.com/huawei-noah/Pretrained-Language-Model/tree/master/NEZHA-TensorFlow/nezha).
+    - sample_text.txt: the file from [google-research/bert](https://github.com/google-research/bert).
+2.  run_read.py: create MindDataset by MindRecord entry script.
+    - create_dataset.py: use MindDataset to read MindRecord to generate dataset.
+
+## Run simple test
+
+Follow the step:
+
+```bash
+bash run.sh         # generate zhwiki.mindrecord* by sample_text.txt
+bash run_read.sh    # use MindDataset to read zhwiki.mindrecord* 
+```
+
+## How to use the example to process zhwiki
+
+Download zhwikidata, extract it, convert it to MindRecord, use MindDataset to read MindRecord.
+
+### Download zhwiki training data
+
+> [zhwiki dataset download address](https://dumps.wikimedia.org/zhwiki) **-> 20200401 -> zhwiki-20200401-pages-articles-multistream.xml.bz2**
+
+### Extract the zhwiki
+
+1. Download [wikiextractor](https://github.com/attardi/wikiextractor) script.
+
+2. Extract the zhwiki.
+    ```python
+    python WikiExtractor.py -o {output_path}/extract {input_path}/zhwiki-20200401-pages-articles-multistream.xml.bz2
+    ```
+
+3. Generate like this:
+    ```
+    $ ls {output_path}/extract
+    AA AB AC AD AE AF AG AH AI AJ AK AL AM AN
+    ```
+
+### Generate MindRecord
+
+1. Modify the parameters in run.sh: --input_file, --output_file, --partition_number.
+    ```
+    --input_file: Input raw text file (or comma-separated list of files).
+    --output_file: Output MindRecord file.
+    --partition_number: The MindRecord file will be split into the number of partition.
+    ```
+
+2. Run the run.sh script.
+    ```
+    bash run.sh
+    ```
+    > Caution: This process is slow, please wait patiently. Run it on server is recommended.
+
+3. The output like this:
+    ```
+    ...
+    [INFO] ME(23485,python):2020-04-28-17:16:40.670.744 [mindspore/ccsrc/mindrecord/io/shard_writer.cc:667] WriteRawData] Write 1 records successfully.
+    [INFO] ME(23485,python):2020-04-28-17:16:40.671.227 [mindspore/ccsrc/mindrecord/io/shard_writer.cc:667] WriteRawData] Write 1 records successfully.
+    [INFO] ME(23485,python):2020-04-28-17:16:40.671.660 [mindspore/ccsrc/mindrecord/io/shard_writer.cc:667] WriteRawData] Write 1 records successfully.
+    [INFO] ME(23485,python):2020-04-28-17:16:40.672.037 [mindspore/ccsrc/mindrecord/io/shard_writer.cc:667] WriteRawData] Write 1 records successfully.
+    [INFO] ME(23485,python):2020-04-28-17:16:40.672.453 [mindspore/ccsrc/mindrecord/io/shard_writer.cc:667] WriteRawData] Write 1 records successfully.
+    [INFO] ME(23485,python):2020-04-28-17:16:40.672.833 [mindspore/ccsrc/mindrecord/io/shard_writer.cc:667] WriteRawData] Write 1 records successfully.
+    ...
+    [INFO] ME(23485:140354285963072,MainProcess):2020-04-28-17:16:40.718.039 [mindspore/mindrecord/filewriter.py:313] The list of mindrecord files created are: ['zhwiki.mindrecord0', 'zhwiki.mindrecord1', 'zhwiki.mindrecord2', 'zhwiki.mindrecord3'], and the list of index files are: ['zhwiki.mindrecord0.db', 'zhwiki.mindrecord1.db', 'zhwiki.mindrecord2.db', 'zhwiki.mindrecord3.db']
+    ...
+    ```
+
+### Create MindDataset By MindRecord
+
+1. Run the run_read.sh script.
+    ```bash
+    bash run_read.sh
+    ```
+
+2. The output like this:
+    ```
+    ...
+    example 74: input_ids: [  101  8168   118 12847  8783  9977 15908   117  8256  9245 11643  8168  8847  8588 11575  8154  8228   143  8384  8376  9197 10241   103 10564 11421  8199 12268   112   161  8228 11541  9586  8436  8174  8363  9864  9702   103   103   119   103  9947 10564   103  8436  8806 11479   103  8912   119   103   103   103 12209  8303   103  8757  8824   117  8256   103  8619  8168 11541   102 11684  8196   103  8228  8847 11523   117  9059  9064 12410  8358  8181 10764   117 11167 11706  9920   148  8332 11390  8936  8205 10951 11997   103  8154   117   103  8670 10467   112   161 10951 13139 12413   117 10288   143 10425  8205   152 10795  8472  8196   103   161 12126  9172 13129 12106  8217  8174 12244  8205   143   103  8461  8277 10628   160  8221   119   102]
+    example 74: input_mask: [1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1]
+    example 74: segment_ids: [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1]
+    example 74: masked_lm_positions: [  6  22  37  38  40  43  47  50  51  52  55  60  67  76  89  92  98 109 120   0]
+    example 74: masked_lm_ids: [ 8118  8165  8329  8890  8554  8458   119  8850  8565 10392  8174 11467  10291  8181  8549 12718 13139   112   158     0]
+    example 74: masked_lm_weights: [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 0.]
+    example 74: next_sentence_labels: [0]
+    ...
+    ```

example/nlp_to_mindrecord/zhwiki/create_dataset.py

@@ -0,0 +1,43 @@
+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# 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.
+# ============================================================================
+"""create MindDataset by MindRecord"""
+import argparse
+import mindspore.dataset as ds
+
+def create_dataset(data_file):
+    """create MindDataset"""
+    num_readers = 4
+    data_set = ds.MindDataset(dataset_file=data_file, num_parallel_workers=num_readers, shuffle=True)
+    index = 0
+    for item in data_set.create_dict_iterator():
+        # print("example {}: {}".format(index, item))
+        print("example {}: input_ids: {}".format(index, item['input_ids']))
+        print("example {}: input_mask: {}".format(index, item['input_mask']))
+        print("example {}: segment_ids: {}".format(index, item['segment_ids']))
+        print("example {}: masked_lm_positions: {}".format(index, item['masked_lm_positions']))
+        print("example {}: masked_lm_ids: {}".format(index, item['masked_lm_ids']))
+        print("example {}: masked_lm_weights: {}".format(index, item['masked_lm_weights']))
+        print("example {}: next_sentence_labels: {}".format(index, item['next_sentence_labels']))
+        index += 1
+        if index % 1000 == 0:
+            print("read rows: {}".format(index))
+    print("total rows: {}".format(index))
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--input_file", type=str, required=True, help='Input mindreord file')
+    args = parser.parse_args()
+
+    create_dataset(args.input_file)

example/nlp_to_mindrecord/zhwiki/create_pretraining_data.py

@@ -0,0 +1,428 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors.
+#
+# 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.
+"""Create masked LM/next sentence masked_lm MindRecord files for BERT."""
+
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import argparse
+import collections
+import logging
+import random
+import tokenization
+
+import numpy as np
+from mindspore.mindrecord import FileWriter
+
+# pylint: skip-file
+
+logging.basicConfig(format='%(asctime)s - %(levelname)s - %(name)s - %(message)s',
+                    datefmt='%m/%d/%Y %H:%M:%S', level=logging.INFO)
+
+
+class TrainingInstance(object):
+  """A single training instance (sentence pair)."""
+
+  def __init__(self, tokens, segment_ids, masked_lm_positions, masked_lm_labels,
+               is_random_next):
+    self.tokens = tokens
+    self.segment_ids = segment_ids
+    self.is_random_next = is_random_next
+    self.masked_lm_positions = masked_lm_positions
+    self.masked_lm_labels = masked_lm_labels
+
+  def __str__(self):
+    s = ""
+    s += "tokens: %s\n" % (" ".join(
+        [tokenization.printable_text(x) for x in self.tokens]))
+    s += "segment_ids: %s\n" % (" ".join([str(x) for x in self.segment_ids]))
+    s += "is_random_next: %s\n" % self.is_random_next
+    s += "masked_lm_positions: %s\n" % (" ".join(
+        [str(x) for x in self.masked_lm_positions]))
+    s += "masked_lm_labels: %s\n" % (" ".join(
+        [tokenization.printable_text(x) for x in self.masked_lm_labels]))
+    s += "\n"
+    return s
+
+  def __repr__(self):
+    return self.__str__()
+
+
+def write_instance_to_example_files(instances, tokenizer, max_seq_length,
+                                    max_predictions_per_seq, output_file, partition_number):
+  """Create MindRecord files from `TrainingInstance`s."""
+  writer = FileWriter(output_file, int(partition_number))
+
+  data_schema = {"input_ids": {"type": "int64", "shape": [-1]},
+                 "input_mask": {"type": "int64", "shape": [-1]},
+                 "segment_ids": {"type": "int64", "shape": [-1]},
+                 "masked_lm_positions": {"type": "int64", "shape": [-1]},
+                 "masked_lm_ids": {"type": "int64", "shape": [-1]},
+                 "masked_lm_weights": {"type": "float64", "shape": [-1]},
+                 "next_sentence_labels": {"type": "int64", "shape": [-1]},
+                }
+  writer.add_schema(data_schema, "zhwiki schema")
+
+  total_written = 0
+  for (inst_index, instance) in enumerate(instances):
+    input_ids = tokenizer.convert_tokens_to_ids(instance.tokens)
+    input_mask = [1] * len(input_ids)
+    segment_ids = list(instance.segment_ids)
+    assert len(input_ids) <= max_seq_length
+
+    while len(input_ids) < max_seq_length:
+      input_ids.append(0)
+      input_mask.append(0)
+      segment_ids.append(0)
+
+    assert len(input_ids) == max_seq_length
+    assert len(input_mask) == max_seq_length
+    assert len(segment_ids) == max_seq_length
+
+    masked_lm_positions = list(instance.masked_lm_positions)
+    masked_lm_ids = tokenizer.convert_tokens_to_ids(instance.masked_lm_labels)
+    masked_lm_weights = [1.0] * len(masked_lm_ids)
+
+    while len(masked_lm_positions) < max_predictions_per_seq:
+      masked_lm_positions.append(0)
+      masked_lm_ids.append(0)
+      masked_lm_weights.append(0.0)
+
+    next_sentence_label = 1 if instance.is_random_next else 0
+
+    features = collections.OrderedDict()
+    features["input_ids"] = np.asarray(input_ids)
+    features["input_mask"] = np.asarray(input_mask)
+    features["segment_ids"] = np.asarray(segment_ids)
+    features["masked_lm_positions"] = np.asarray(masked_lm_positions)
+    features["masked_lm_ids"] = np.asarray(masked_lm_ids)
+    features["masked_lm_weights"] = np.asarray(masked_lm_weights)
+    features["next_sentence_labels"] = np.asarray([next_sentence_label])
+
+    total_written += 1
+
+    if inst_index < 20:
+      logging.info("*** Example ***")
+      logging.info("tokens: %s" % " ".join(
+          [tokenization.printable_text(x) for x in instance.tokens]))
+
+      for feature_name in features.keys():
+        feature = features[feature_name]
+        logging.info(
+            "%s: %s" % (feature_name, " ".join([str(x) for x in feature])))
+    writer.write_raw_data([features])
+
+  writer.commit()
+
+  logging.info("Wrote %d total instances", total_written)
+
+
+def create_training_instances(input_files, tokenizer, max_seq_length,
+                              dupe_factor, short_seq_prob, masked_lm_prob,
+                              max_predictions_per_seq, rng, do_whole_word_mask):
+  """Create `TrainingInstance`s from raw text."""
+  all_documents = [[]]
+
+  # Input file format:
+  # (1) One sentence per line. These should ideally be actual sentences, not
+  # entire paragraphs or arbitrary spans of text. (Because we use the
+  # sentence boundaries for the "next sentence prediction" task).
+  # (2) Blank lines between documents. Document boundaries are needed so
+  # that the "next sentence prediction" task doesn't span between documents.
+  for input_file in input_files:
+    with open(input_file, "r") as reader:
+      while True:
+        line = tokenization.convert_to_unicode(reader.readline())
+        if not line:
+          break
+        line = line.strip()
+
+        # Empty lines are used as document delimiters
+        if not line:
+          all_documents.append([])
+        tokens = tokenizer.tokenize(line)
+        if tokens:
+          all_documents[-1].append(tokens)
+
+  # Remove empty documents
+  all_documents = [x for x in all_documents if x]
+  rng.shuffle(all_documents)
+
+  vocab_words = list(tokenizer.vocab.keys())
+  instances = []
+  for _ in range(dupe_factor):
+    for document_index in range(len(all_documents)):
+      instances.extend(
+          create_instances_from_document(
+              all_documents, document_index, max_seq_length, short_seq_prob,
+              masked_lm_prob, max_predictions_per_seq, vocab_words, rng, do_whole_word_mask))
+
+  rng.shuffle(instances)
+  return instances
+
+
+def create_instances_from_document(
+    all_documents, document_index, max_seq_length, short_seq_prob,
+    masked_lm_prob, max_predictions_per_seq, vocab_words, rng, do_whole_word_mask):
+  """Creates `TrainingInstance`s for a single document."""
+  document = all_documents[document_index]
+
+  # Account for [CLS], [SEP], [SEP]
+  max_num_tokens = max_seq_length - 3
+
+  # We *usually* want to fill up the entire sequence since we are padding
+  # to `max_seq_length` anyways, so short sequences are generally wasted
+  # computation. However, we *sometimes*
+  # (i.e., short_seq_prob == 0.1 == 10% of the time) want to use shorter
+  # sequences to minimize the mismatch between pre-training and fine-tuning.
+  # The `target_seq_length` is just a rough target however, whereas
+  # `max_seq_length` is a hard limit.
+  target_seq_length = max_num_tokens
+  if rng.random() < short_seq_prob:
+    target_seq_length = rng.randint(2, max_num_tokens)
+
+  # We DON'T just concatenate all of the tokens from a document into a long
+  # sequence and choose an arbitrary split point because this would make the
+  # next sentence prediction task too easy. Instead, we split the input into
+  # segments "A" and "B" based on the actual "sentences" provided by the user
+  # input.
+  instances = []
+  current_chunk = []
+  current_length = 0
+  i = 0
+  while i < len(document):
+    segment = document[i]
+    current_chunk.append(segment)
+    current_length += len(segment)
+    if i == len(document) - 1 or current_length >= target_seq_length:
+      if current_chunk:
+        # `a_end` is how many segments from `current_chunk` go into the `A`
+        # (first) sentence.
+        a_end = 1
+        if len(current_chunk) >= 2:
+          a_end = rng.randint(1, len(current_chunk) - 1)
+
+        tokens_a = []
+        for j in range(a_end):
+          tokens_a.extend(current_chunk[j])
+
+        tokens_b = []
+        # Random next
+        is_random_next = False
+        if len(current_chunk) == 1 or rng.random() < 0.5:
+          is_random_next = True
+          target_b_length = target_seq_length - len(tokens_a)
+
+          # This should rarely go for more than one iteration for large
+          # corpora. However, just to be careful, we try to make sure that
+          # the random document is not the same as the document
+          # we're processing.
+          for _ in range(10):
+            random_document_index = rng.randint(0, len(all_documents) - 1)
+            if random_document_index != document_index:
+              break
+
+          random_document = all_documents[random_document_index]
+          random_start = rng.randint(0, len(random_document) - 1)
+          for j in range(random_start, len(random_document)):
+            tokens_b.extend(random_document[j])
+            if len(tokens_b) >= target_b_length:
+              break
+          # We didn't actually use these segments so we "put them back" so
+          # they don't go to waste.
+          num_unused_segments = len(current_chunk) - a_end
+          i -= num_unused_segments
+        # Actual next
+        else:
+          is_random_next = False
+          for j in range(a_end, len(current_chunk)):
+            tokens_b.extend(current_chunk[j])
+        truncate_seq_pair(tokens_a, tokens_b, max_num_tokens, rng)
+
+        assert len(tokens_a) >= 1
+        assert len(tokens_b) >= 1
+
+        tokens = []
+        segment_ids = []
+        tokens.append("[CLS]")
+        segment_ids.append(0)
+        for token in tokens_a:
+          tokens.append(token)
+          segment_ids.append(0)
+
+        tokens.append("[SEP]")
+        segment_ids.append(0)
+
+        for token in tokens_b:
+          tokens.append(token)
+          segment_ids.append(1)
+        tokens.append("[SEP]")
+        segment_ids.append(1)
+
+        (tokens, masked_lm_positions,
+         masked_lm_labels) = create_masked_lm_predictions(
+             tokens, masked_lm_prob, max_predictions_per_seq, vocab_words, rng, do_whole_word_mask)
+        instance = TrainingInstance(
+            tokens=tokens,
+            segment_ids=segment_ids,
+            is_random_next=is_random_next,
+            masked_lm_positions=masked_lm_positions,
+            masked_lm_labels=masked_lm_labels)
+        instances.append(instance)
+      current_chunk = []
+      current_length = 0
+    i += 1
+
+  return instances
+
+
+MaskedLmInstance = collections.namedtuple("MaskedLmInstance",
+                                          ["index", "label"])
+
+
+def create_masked_lm_predictions(tokens, masked_lm_prob,
+                                 max_predictions_per_seq, vocab_words, rng, do_whole_word_mask):
+  """Creates the predictions for the masked LM objective."""
+
+  cand_indexes = []
+  for (i, token) in enumerate(tokens):
+    if token == "[CLS]" or token == "[SEP]":
+      continue
+    # Whole Word Masking means that if we mask all of the wordpieces
+    # corresponding to an original word. When a word has been split into
+    # WordPieces, the first token does not have any marker and any subsequence
+    # tokens are prefixed with ##. So whenever we see the ## token, we
+    # append it to the previous set of word indexes.
+    #
+    # Note that Whole Word Masking does *not* change the training code
+    # at all -- we still predict each WordPiece independently, softmaxed
+    # over the entire vocabulary.
+    if (do_whole_word_mask and len(cand_indexes) >= 1 and
+        token.startswith("##")):
+      cand_indexes[-1].append(i)
+    else:
+      cand_indexes.append([i])
+
+  rng.shuffle(cand_indexes)
+
+  output_tokens = list(tokens)
+
+  num_to_predict = min(max_predictions_per_seq,
+                       max(1, int(round(len(tokens) * masked_lm_prob))))
+
+  masked_lms = []
+  covered_indexes = set()
+  for index_set in cand_indexes:
+    if len(masked_lms) >= num_to_predict:
+      break
+    # If adding a whole-word mask would exceed the maximum number of
+    # predictions, then just skip this candidate.
+    if len(masked_lms) + len(index_set) > num_to_predict:
+      continue
+    is_any_index_covered = False
+    for index in index_set:
+      if index in covered_indexes:
+        is_any_index_covered = True
+        break
+    if is_any_index_covered:
+      continue
+    for index in index_set:
+      covered_indexes.add(index)
+
+      masked_token = None
+      # 80% of the time, replace with [MASK]
+      if rng.random() < 0.8:
+        masked_token = "[MASK]"
+      else:
+        # 10% of the time, keep original
+        if rng.random() < 0.5:
+          masked_token = tokens[index]
+        # 10% of the time, replace with random word
+        else:
+          masked_token = vocab_words[rng.randint(0, len(vocab_words) - 1)]
+
+      output_tokens[index] = masked_token
+
+      masked_lms.append(MaskedLmInstance(index=index, label=tokens[index]))
+  assert len(masked_lms) <= num_to_predict
+  masked_lms = sorted(masked_lms, key=lambda x: x.index)
+
+  masked_lm_positions = []
+  masked_lm_labels = []
+  for p in masked_lms:
+    masked_lm_positions.append(p.index)
+    masked_lm_labels.append(p.label)
+
+  return (output_tokens, masked_lm_positions, masked_lm_labels)
+
+
+def truncate_seq_pair(tokens_a, tokens_b, max_num_tokens, rng):
+  """Truncates a pair of sequences to a maximum sequence length."""
+  while True:
+    total_length = len(tokens_a) + len(tokens_b)
+    if total_length <= max_num_tokens:
+      break
+
+    trunc_tokens = tokens_a if len(tokens_a) > len(tokens_b) else tokens_b
+    assert len(trunc_tokens) >= 1
+
+    # We want to sometimes truncate from the front and sometimes from the
+    # back to add more randomness and avoid biases.
+    if rng.random() < 0.5:
+      del trunc_tokens[0]
+    else:
+      trunc_tokens.pop()
+
+
+def main():
+  parser = argparse.ArgumentParser()
+  parser.add_argument("--input_file", type=str, required=True, help='Input raw text file (or comma-separated list of files).')
+  parser.add_argument("--output_file", type=str, required=True, help='Output MindRecord file.')
+  parser.add_argument("--partition_number", type=int, default=1, help='The MindRecord file will be split into the number of partition.')
+  parser.add_argument("--vocab_file", type=str, required=True, help='The vocabulary file than the BERT model was trained on.')
+  parser.add_argument("--do_lower_case", type=bool, default=False, help='Whether to lower case the input text. Should be True for uncased models and False for cased models.')
+  parser.add_argument("--do_whole_word_mask", type=bool, default=False, help='Whether to use whole word masking rather than per-WordPiece masking.')
+  parser.add_argument("--max_seq_length", type=int, default=128, help='Maximum sequence length.')
+  parser.add_argument("--max_predictions_per_seq", type=int, default=20, help='Maximum number of masked LM predictions per sequence.')
+  parser.add_argument("--random_seed", type=int, default=12345, help='Random seed for data generation.')
+  parser.add_argument("--dupe_factor", type=int, default=10, help='Number of times to duplicate the input data (with diffrent masks).')
+  parser.add_argument("--masked_lm_prob", type=float, default=0.15, help='Masked LM probability.')
+  parser.add_argument("--short_seq_prob", type=float, default=0.1, help='Probability of creating sequences which are shorter than the maximum length.')
+  args = parser.parse_args()
+
+  tokenizer = tokenization.FullTokenizer(
+      vocab_file=args.vocab_file, do_lower_case=args.do_lower_case)
+
+  input_files = []
+  for input_pattern in args.input_file.split(","):
+    input_files.append(input_pattern)
+
+  logging.info("*** Reading from input files ***")
+  for input_file in input_files:
+    logging.info("  %s", input_file)
+
+  rng = random.Random(args.random_seed)
+  instances = create_training_instances(
+      input_files, tokenizer, args.max_seq_length, args.dupe_factor,
+      args.short_seq_prob, args.masked_lm_prob, args.max_predictions_per_seq,
+      rng, args.do_whole_word_mask)
+
+  write_instance_to_example_files(instances, tokenizer, args.max_seq_length,
+                                  args.max_predictions_per_seq, args.output_file, args.partition_number)
+
+
+if __name__ == "__main__":
+  main()

example/nlp_to_mindrecord/zhwiki/run.sh

@@ -0,0 +1,29 @@
+#!/bin/bash
+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# 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.
+# ============================================================================
+
+rm zhwiki.mindrecord*
+
+python create_pretraining_data.py \
+--input_file=./sample_text.txt \
+--output_file=zhwiki.mindrecord \
+--partition_number=4 \
+--vocab_file=./vocab.txt \
+--do_lower_case=True \
+--max_seq_length=128 \
+--max_predictions_per_seq=20 \
+--masked_lm_prob=0.15 \
+--random_seed=12345 \
+--dupe_factor=5

example/nlp_to_mindrecord/zhwiki/run_read.sh

@@ -0,0 +1,17 @@
+#!/bin/bash
+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# 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.
+# ============================================================================
+
+python create_dataset.py --input_file=zhwiki.mindrecord0

example/nlp_to_mindrecord/zhwiki/sample_text.txt

@@ -0,0 +1,33 @@
+This text is included to make sure Unicode is handled properly: 力加勝北区ᴵᴺᵀᵃছজটডণত
+Text should be one-sentence-per-line, with empty lines between documents.
+This sample text is public domain and was randomly selected from Project Guttenberg.
+
+The rain had only ceased with the gray streaks of morning at Blazing Star, and the settlement awoke to a moral sense of cleanliness, and the finding of forgotten knives, tin cups, and smaller camp utensils, where the heavy showers had washed away the debris and dust heaps before the cabin doors.
+Indeed, it was recorded in Blazing Star that a fortunate early riser had once picked up on the highway a solid chunk of gold quartz which the rain had freed from its incumbering soil, and washed into immediate and glittering popularity.
+Possibly this may have been the reason why early risers in that locality, during the rainy season, adopted a thoughtful habit of body, and seldom lifted their eyes to the rifted or india-ink washed skies above them.
+"Cass" Beard had risen early that morning, but not with a view to discovery.
+A leak in his cabin roof,--quite consistent with his careless, improvident habits,--had roused him at 4 A. M., with a flooded "bunk" and wet blankets.
+The chips from his wood pile refused to kindle a fire to dry his bed-clothes, and he had recourse to a more provident neighbor's to supply the deficiency.
+This was nearly opposite.
+Mr. Cassius crossed the highway, and stopped suddenly.
+Something glittered in the nearest red pool before him.
+Gold, surely!
+But, wonderful to relate, not an irregular, shapeless fragment of crude ore, fresh from Nature's crucible, but a bit of jeweler's handicraft in the form of a plain gold ring.
+Looking at it more attentively, he saw that it bore the inscription, "May to Cass."
+Like most of his fellow gold-seekers, Cass was superstitious.
+
+The fountain of classic wisdom, Hypatia herself.
+As the ancient sage--the name is unimportant to a monk--pumped water nightly that he might study by day, so I, the guardian of cloaks and parasols, at the sacred doors of her lecture-room, imbibe celestial knowledge.
+From my youth I felt in me a soul above the matter-entangled herd.
+She revealed to me the glorious fact, that I am a spark of Divinity itself.
+A fallen star, I am, sir!' continued he, pensively, stroking his lean stomach--'a fallen star!--fallen, if the dignity of philosophy will allow of the simile, among the hogs of the lower world--indeed, even into the hog-bucket itself. Well, after all, I will show you the way to the Archbishop's.
+There is a philosophic pleasure in opening one's treasures to the modest young.
+Perhaps you will assist me by carrying this basket of fruit?' And the little man jumped up, put his basket on Philammon's head, and trotted off up a neighbouring street.
+Philammon followed, half contemptuous, half wondering at what this philosophy might be, which could feed the self-conceit of anything so abject as his ragged little apish guide;
+but the novel roar and whirl of the street, the perpetual stream of busy faces, the line of curricles, palanquins, laden asses, camels, elephants, which met and passed him, and squeezed him up steps and into doorways, as they threaded their way through the great Moon-gate into the ample street beyond, drove everything from his mind but wondering curiosity, and a vague, helpless dread of that great living wilderness, more terrible than any dead wilderness of sand which he had left behind.
+Already he longed for the repose, the silence of the Laura--for faces which knew him and smiled upon him; but it was too late to turn back now.
+His guide held on for more than a mile up the great main street, crossed in the centre of the city, at right angles, by one equally magnificent, at each end of which, miles away, appeared, dim and distant over the heads of the living stream of passengers, the yellow sand-hills of the desert;
+while at the end of the vista in front of them gleamed the blue harbour, through a network of countless masts.
+At last they reached the quay at the opposite end of the street;
+and there burst on Philammon's astonished eyes a vast semicircle of blue sea, ringed with palaces and towers.
+He stopped involuntarily; and his little guide stopped also, and looked askance at the young monk, to watch the effect which that grand panorama should produce on him.

example/nlp_to_mindrecord/zhwiki/tokenization.py

@@ -0,0 +1,394 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors.
+#
+# 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.
+"""Tokenization classes."""
+
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import collections
+import re
+import unicodedata
+import six
+
+# pylint: skip-file
+
+def validate_case_matches_checkpoint(do_lower_case, init_checkpoint):
+  """Checks whether the casing config is consistent with the checkpoint name."""
+
+  # The casing has to be passed in by the user and there is no explicit check
+  # as to whether it matches the checkpoint. The casing information probably
+  # should have been stored in the bert_config.json file, but it's not, so
+  # we have to heuristically detect it to validate.
+
+  if not init_checkpoint:
+    return
+
+  m = re.match("^.*?([A-Za-z0-9_-]+)/bert_model.ckpt", init_checkpoint)
+  if m is None:
+    return
+
+  model_name = m.group(1)
+
+  lower_models = [
+      "uncased_L-24_H-1024_A-16", "uncased_L-12_H-768_A-12",
+      "multilingual_L-12_H-768_A-12", "chinese_L-12_H-768_A-12"
+  ]
+
+  cased_models = [
+      "cased_L-12_H-768_A-12", "cased_L-24_H-1024_A-16",
+      "multi_cased_L-12_H-768_A-12"
+  ]
+
+  is_bad_config = False
+  if model_name in lower_models and not do_lower_case:
+    is_bad_config = True
+    actual_flag = "False"
+    case_name = "lowercased"
+    opposite_flag = "True"
+
+  if model_name in cased_models and do_lower_case:
+    is_bad_config = True
+    actual_flag = "True"
+    case_name = "cased"
+    opposite_flag = "False"
+
+  if is_bad_config:
+    raise ValueError(
+        "You passed in `--do_lower_case=%s` with `--init_checkpoint=%s`. "
+        "However, `%s` seems to be a %s model, so you "
+        "should pass in `--do_lower_case=%s` so that the fine-tuning matches "
+        "how the model was pre-training. If this error is wrong, please "
+        "just comment out this check." % (actual_flag, init_checkpoint,
+                                          model_name, case_name, opposite_flag))
+
+
+def convert_to_unicode(text):
+  """Converts `text` to Unicode (if it's not already), assuming utf-8 input."""
+  if six.PY3:
+    if isinstance(text, str):
+      return text
+    elif isinstance(text, bytes):
+      return text.decode("utf-8", "ignore")
+    else:
+      raise ValueError("Unsupported string type: %s" % (type(text)))
+  elif six.PY2:
+    if isinstance(text, str):
+      return text.decode("utf-8", "ignore")
+    elif isinstance(text, unicode):
+      return text
+    else:
+      raise ValueError("Unsupported string type: %s" % (type(text)))
+  else:
+    raise ValueError("Not running on Python2 or Python 3?")
+
+
+def printable_text(text):
+  """Returns text encoded in a way suitable for print or `tf.logging`."""
+
+  # These functions want `str` for both Python2 and Python3, but in one case
+  # it's a Unicode string and in the other it's a byte string.
+  if six.PY3:
+    if isinstance(text, str):
+      return text
+    elif isinstance(text, bytes):
+      return text.decode("utf-8", "ignore")
+    else:
+      raise ValueError("Unsupported string type: %s" % (type(text)))
+  elif six.PY2:
+    if isinstance(text, str):
+      return text
+    elif isinstance(text, unicode):
+      return text.encode("utf-8")
+    else:
+      raise ValueError("Unsupported string type: %s" % (type(text)))
+  else:
+    raise ValueError("Not running on Python2 or Python 3?")
+
+
+def load_vocab(vocab_file):
+  """Loads a vocabulary file into a dictionary."""
+  vocab = collections.OrderedDict()
+  index = 0
+  with open(vocab_file, "r") as reader:
+    while True:
+      token = convert_to_unicode(reader.readline())
+      if not token:
+        break
+      token = token.strip()
+      vocab[token] = index
+      index += 1
+  return vocab
+
+
+def convert_by_vocab(vocab, items):
+  """Converts a sequence of [tokens|ids] using the vocab."""
+  output = []
+  for item in items:
+    output.append(vocab[item])
+  return output
+
+
+def convert_tokens_to_ids(vocab, tokens):
+  return convert_by_vocab(vocab, tokens)
+
+
+def convert_ids_to_tokens(inv_vocab, ids):
+  return convert_by_vocab(inv_vocab, ids)
+
+
+def whitespace_tokenize(text):
+  """Runs basic whitespace cleaning and splitting on a piece of text."""
+  text = text.strip()
+  if not text:
+    return []
+  tokens = text.split()
+  return tokens
+
+
+class FullTokenizer(object):
+  """Runs end-to-end tokenziation."""
+
+  def __init__(self, vocab_file, do_lower_case=True):
+    self.vocab = load_vocab(vocab_file)
+    self.inv_vocab = {v: k for k, v in self.vocab.items()}
+    self.basic_tokenizer = BasicTokenizer(do_lower_case=do_lower_case)
+    self.wordpiece_tokenizer = WordpieceTokenizer(vocab=self.vocab)
+
+  def tokenize(self, text):
+    split_tokens = []
+    for token in self.basic_tokenizer.tokenize(text):
+      for sub_token in self.wordpiece_tokenizer.tokenize(token):
+        split_tokens.append(sub_token)
+
+    return split_tokens
+
+  def convert_tokens_to_ids(self, tokens):
+    return convert_by_vocab(self.vocab, tokens)
+
+  def convert_ids_to_tokens(self, ids):
+    return convert_by_vocab(self.inv_vocab, ids)
+
+
+class BasicTokenizer(object):
+  """Runs basic tokenization (punctuation splitting, lower casing, etc.)."""
+
+  def __init__(self, do_lower_case=True):
+    """Constructs a BasicTokenizer.
+    Args:
+      do_lower_case: Whether to lower case the input.
+    """
+    self.do_lower_case = do_lower_case
+
+  def tokenize(self, text):
+    """Tokenizes a piece of text."""
+    text = convert_to_unicode(text)
+    text = self._clean_text(text)
+
+    # This was added on November 1st, 2018 for the multilingual and Chinese
+    # models. This is also applied to the English models now, but it doesn't
+    # matter since the English models were not trained on any Chinese data
+    # and generally don't have any Chinese data in them (there are Chinese
+    # characters in the vocabulary because Wikipedia does have some Chinese
+    # words in the English Wikipedia.).
+    text = self._tokenize_chinese_chars(text)
+
+    orig_tokens = whitespace_tokenize(text)
+    split_tokens = []
+    for token in orig_tokens:
+      if self.do_lower_case:
+        token = token.lower()
+        token = self._run_strip_accents(token)
+      split_tokens.extend(self._run_split_on_punc(token))
+
+    output_tokens = whitespace_tokenize(" ".join(split_tokens))
+    return output_tokens
+
+  def _run_strip_accents(self, text):
+    """Strips accents from a piece of text."""
+    text = unicodedata.normalize("NFD", text)
+    output = []
+    for char in text:
+      cat = unicodedata.category(char)
+      if cat == "Mn":
+        continue
+      output.append(char)
+    return "".join(output)
+
+  def _run_split_on_punc(self, text):
+    """Splits punctuation on a piece of text."""
+    chars = list(text)
+    i = 0
+    start_new_word = True
+    output = []
+    while i < len(chars):
+      char = chars[i]
+      if _is_punctuation(char):
+        output.append([char])
+        start_new_word = True
+      else:
+        if start_new_word:
+          output.append([])
+        start_new_word = False
+        output[-1].append(char)
+      i += 1
+
+    return ["".join(x) for x in output]
+
+  def _tokenize_chinese_chars(self, text):
+    """Adds whitespace around any CJK character."""
+    output = []
+    for char in text:
+      cp = ord(char)
+      if self._is_chinese_char(cp):
+        output.append(" ")
+        output.append(char)
+        output.append(" ")
+      else:
+        output.append(char)
+    return "".join(output)
+
+  def _is_chinese_char(self, cp):
+    """Checks whether CP is the codepoint of a CJK character."""
+    # This defines a "chinese character" as anything in the CJK Unicode block:
+    #   https://en.wikipedia.org/wiki/CJK_Unified_Ideographs_(Unicode_block)
+    #
+    # Note that the CJK Unicode block is NOT all Japanese and Korean characters,
+    # despite its name. The modern Korean Hangul alphabet is a different block,
+    # as is Japanese Hiragana and Katakana. Those alphabets are used to write
+    # space-separated words, so they are not treated specially and handled
+    # like the all of the other languages.
+    if ((cp >= 0x4E00 and cp <= 0x9FFF) or  #
+        (cp >= 0x3400 and cp <= 0x4DBF) or  #
+        (cp >= 0x20000 and cp <= 0x2A6DF) or  #
+        (cp >= 0x2A700 and cp <= 0x2B73F) or  #
+        (cp >= 0x2B740 and cp <= 0x2B81F) or  #
+        (cp >= 0x2B820 and cp <= 0x2CEAF) or
+        (cp >= 0xF900 and cp <= 0xFAFF) or  #
+        (cp >= 0x2F800 and cp <= 0x2FA1F)):  #
+      return True
+
+    return False
+
+  def _clean_text(self, text):
+    """Performs invalid character removal and whitespace cleanup on text."""
+    output = []
+    for char in text:
+      cp = ord(char)
+      if cp == 0 or cp == 0xfffd or _is_control(char):
+        continue
+      if _is_whitespace(char):
+        output.append(" ")
+      else:
+        output.append(char)
+    return "".join(output)
+
+
+class WordpieceTokenizer(object):
+  """Runs WordPiece tokenziation."""
+
+  def __init__(self, vocab, unk_token="[UNK]", max_input_chars_per_word=200):
+    self.vocab = vocab
+    self.unk_token = unk_token
+    self.max_input_chars_per_word = max_input_chars_per_word
+
+  def tokenize(self, text):
+    """Tokenizes a piece of text into its word pieces.
+    This uses a greedy longest-match-first algorithm to perform tokenization
+    using the given vocabulary.
+    For example:
+      input = "unaffable"
+      output = ["un", "##aff", "##able"]
+    Args:
+      text: A single token or whitespace separated tokens. This should have
+        already been passed through `BasicTokenizer.
+    Returns:
+      A list of wordpiece tokens.
+    """
+
+    text = convert_to_unicode(text)
+
+    output_tokens = []
+    for token in whitespace_tokenize(text):
+      chars = list(token)
+      if len(chars) > self.max_input_chars_per_word:
+        output_tokens.append(self.unk_token)
+        continue
+
+      is_bad = False
+      start = 0
+      sub_tokens = []
+      while start < len(chars):
+        end = len(chars)
+        cur_substr = None
+        while start < end:
+          substr = "".join(chars[start:end])
+          if start > 0:
+            substr = "##" + substr
+          if substr in self.vocab:
+            cur_substr = substr
+            break
+          end -= 1
+        if cur_substr is None:
+          is_bad = True
+          break
+        sub_tokens.append(cur_substr)
+        start = end
+
+      if is_bad:
+        output_tokens.append(self.unk_token)
+      else:
+        output_tokens.extend(sub_tokens)
+    return output_tokens
+
+
+def _is_whitespace(char):
+  """Checks whether `chars` is a whitespace character."""
+  # \t, \n, and \r are technically contorl characters but we treat them
+  # as whitespace since they are generally considered as such.
+  if char == " " or char == "\t" or char == "\n" or char == "\r":
+    return True
+  cat = unicodedata.category(char)
+  if cat == "Zs":
+    return True
+  return False
+
+
+def _is_control(char):
+  """Checks whether `chars` is a control character."""
+  # These are technically control characters but we count them as whitespace
+  # characters.
+  if char == "\t" or char == "\n" or char == "\r":
+    return False
+  cat = unicodedata.category(char)
+  if cat in ("Cc", "Cf"):
+    return True
+  return False
+
+
+def _is_punctuation(char):
+  """Checks whether `chars` is a punctuation character."""
+  cp = ord(char)
+  # We treat all non-letter/number ASCII as punctuation.
+  # Characters such as "^", "$", and "`" are not in the Unicode
+  # Punctuation class but we treat them as punctuation anyways, for
+  # consistency.
+  if ((cp >= 33 and cp <= 47) or (cp >= 58 and cp <= 64) or
+      (cp >= 91 and cp <= 96) or (cp >= 123 and cp <= 126)):
+    return True
+  cat = unicodedata.category(char)
+  if cat.startswith("P"):
+    return True
+  return False