bert_classify.py

# -*- coding: utf-8 -*-
# @Author: gunjianpan
# @Date:   2019-05-21 11:21:42
# @Last Modified by:   gunjianpan
# @Last Modified time: 2019-05-21 11:25:01

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import collections
import os
import numpy as np
import pickle
import bert.modeling as modeling
import bert.optimization as optimization
import bert.tokenization as tokenization
import tensorflow as tf

from sklearn.model_selection import train_test_split
from constant import *
from util import load_embedding, pad_middle, load_result_f1, time_str
from data_load import SemEvalDataLoader

flags = tf.flags

FLAGS = flags.FLAGS
tf.data.experimental.ignore_errors()

bert_dir = '/Users/gunjianpan/Desktop/git/bert/uncased_L-24_H-1024_A-16/'

''' for file path '''
flags.DEFINE_string("task_name", 'SemEval2017Task4', "task name")
flags.DEFINE_string("data_dir", './', 'data dir')
flags.DEFINE_string("output_dir", './result/', "output dir")
flags.DEFINE_string("vocab_file", '{}vocab.txt'.format(bert_dir), "")
flags.DEFINE_string("bert_config_file",'{}bert_config.json'.format(bert_dir), "pretrain configure model")
flags.DEFINE_string("init_checkpoint",'{}bert_model.ckpt'.format(bert_dir), "")

''' for pattern '''
flags.DEFINE_bool("do_lower_case", True, "if low word")

flags.DEFINE_bool("do_train", False, "Whether to run training.")
flags.DEFINE_bool("do_eval", False, "Whether to run eval on the dev set.")
flags.DEFINE_bool("do_predict", True, "whether to run predict")

tf.app.flags.DEFINE_integer("pad_type", 0, "pad type")
tf.app.flags.DEFINE_boolean("ekphrasis", True, "ekphrasis type")

''' for hyper-parameters '''
flags.DEFINE_integer("train_batch_size", 32, "Total batch size for training.")
flags.DEFINE_integer("eval_batch_size", 8, "Total batch size for eval.")
flags.DEFINE_integer("predict_batch_size", 8, "Total batch size for predict.")

flags.DEFINE_integer("max_seq_length", 128, "truncated/padded")
flags.DEFINE_float("learning_rate", 5e-5, "learning rate Adam.")
flags.DEFINE_float("num_train_epochs", 2.0, "train epochs")
flags.DEFINE_float("warmup_proportion", 0.1, "warming proportion")

flags.DEFINE_integer("save_checkpoints_steps", 1000, "")

''' for TPU'''
flags.DEFINE_integer("iterations_per_loop", 1000, "for TPU")
flags.DEFINE_bool("use_tpu", False, "Whether to use TPU or GPU/CPU.")
tf.flags.DEFINE_string("tpu_name", None, "")
tf.flags.DEFINE_string("tpu_zone", None, "")
tf.flags.DEFINE_string("gcp_project", None, "")
tf.flags.DEFINE_string("master", None, "[Optional] TensorFlow master URL.")

flags.DEFINE_integer("num_tpu_cores", 8, "")


class InputClass(object):
    ''' Input class '''

    def __init__(self, guid, text_a, text_b=None, label=None):
        """Constructs a InputClass.

        Args:
          guid: Unique id for the example.
          text_a: string. The untokenized text of the first sequence. For single
            sequence tasks, only this sequence must be specified.
          text_b: (Optional) string. The untokenized text of the second sequence.
            Only must be specified for sequence pair tasks.
          label: (Optional) string. The label of the example. This should be
            specified for train and dev examples, but not for test examples.
        """
        self.guid = guid
        self.text_a = text_a
        self.text_b = text_b
        self.label = label


class PaddingClass(object):
    """Fake example so the num input examples is a multiple of the batch size.

    When running eval/predict on the TPU, we need to pad the number of examples
    to be a multiple of the batch size, because the TPU requires a fixed batch
    size. The alternative is to drop the last batch, which is bad because it means
    the entire output data won't be generated.

    We use this class instead of `None` because treating `None` as padding
    battches could cause silent errors.
    """


class InputFeatures(object):
    """A single set of features of data."""

    def __init__(self,
                 input_ids,
                 input_mask,
                 segment_ids,
                 label_id,
                 is_real_example=True):
        self.input_ids = input_ids
        self.input_mask = input_mask
        self.segment_ids = segment_ids
        self.label_id = label_id
        self.is_real_example = is_real_example


class SemEval2017Task4Processor(DataProcessor):
    """ SemEval 2017 Task 4 subTask 4 Processor (Multi-class) """

    def __init__(self):
        self.load_data()

    def get_train_examples(self, data_dir):
        return self.train

    def get_dev_examples(self, data_dir):
        return self.dev

    def get_test_examples(self, data_dir):
        return self.test

    def get_labels(self):
        return list(range(num_class))

    def load_data(self):
        data_path = '{}data_{}.pkl'.format(pickle_dir, FLAGS.ekphrasis)
        if os.path.exists(data_path):
            print(11111)
            train_set, test_data = pickle.load(open(data_path, 'rb'))
        else:
            train_set = SemEvalDataLoader(verbose=False, ekphrasis=FLAGS.ekphrasis).get_data(task="A",
                                                                                             years=None,
                                                                                             datasets=None,
                                                                                             only_semEval=True)
            test_data = SemEvalDataLoader(
                verbose=False, ekphrasis=FLAGS.ekphrasis).get_gold(task="A")
            pickle.dump([train_set, test_data], open(data_path, 'wb'))
        X = [obs[1] for obs in train_set]
        y = [label2id[obs[0]] for obs in train_set]
        X_test = [obs[1] for obs in test_data]
        test_Y = [label2id[obs[0]] for obs in test_data]

        sentences_len = [len(ii.split()) for ii in [*X, *X_test]]
        sent_size = max(sentences_len)
        pad_type = FLAGS.pad_type

        sent_re = [tokenization.convert_to_unicode(ii) for ii in X]
        test_sent_out = [tokenization.convert_to_unicode(ii)for ii in X_test]

        train_X, X_test, train_Y, y_test = train_test_split(
            sent_re, y, test_size=0.25)
        train = [InputClass(guid='train-{}'.format(ii), text_a=tokenization.convert_to_unicode(
            jj), text_b=None, label=train_Y[ii]) for ii, jj in enumerate(train_X)]
        dev = [InputClass(guid='dev-{}'.format(ii), text_a=tokenization.convert_to_unicode(
            jj), text_b=None, label=y_test[ii]) for ii, jj in enumerate(X_test)]
        test = [InputClass(guid='test-{}'.format(ii), text_a=tokenization.convert_to_unicode(
            jj), text_b=None, label=test_Y[ii]) for ii, jj in enumerate(test_sent_out)]
        self.train = train
        self.dev = dev
        self.test = test


def convert_single_example(ex_index, example, label_list, max_seq_length, tokenizer):
    '''Converts a single `InputClass` into a single `InputFeatures`.'''

    if isinstance(example, PaddingClass):
        return InputFeatures(
            input_ids=[0] * max_seq_length,
            input_mask=[0] * max_seq_length,
            segment_ids=[0] * max_seq_length,
            label_id=0,
            is_real_example=False)

    label_map = {label: i for (i, label) in enumerate(label_list)}

    tokens_a = tokenizer.tokenize(example.text_a)
    tokens_b = tokenizer.tokenize(example.text_b) if example.text_b else None

    if tokens_b:  # opt. for no take_b, - 3 = [CLS] + [SEP] * 2
        _truncate_seq_pair(tokens_a, tokens_b, max_seq_length - 3)
    else:  # - 2 = [CLS] + [SEP]
        if len(tokens_a) > max_seq_length - 2:
            tokens_a = tokens_a[0:(max_seq_length - 2)]

    # (a) For sequence pairs:
    #  tokens:   [CLS] is this jack ##son ##ville ? [SEP] no it is not . [SEP]
    #  type_ids: 0     0  0    0    0     0       0 0     1  1  1  1   1 1
    # (b) For single sequences:
    #  tokens:   [CLS] the dog is hairy . [SEP]
    #  type_ids: 0     0   0   0  0     0 0
    tokens = ['[CLS]', *[ii for ii in tokens_a], '[SEP]']
    segment_ids = [0] * (len(tokens_a) + 1)  # No.X sentences

    if tokens_b:
        tokens = [*tokens, *[ii for ii in tokens_b], '[SEP]']
        segment_ids = [*segment_ids, *[1] * (len(tokens_b) + 1)]

    input_ids = tokenizer.convert_tokens_to_ids(tokens)

    input_mask = [1] * len(input_ids)  # mask 0: pad; 1: real

    pad_len = max_seq_length - len(input_ids)
    if not pad_len:  # zero-padded
        input_ids = [*input_ids, *[0] * pad_len]
        input_mask = [*input_mask, *[0] * pad_len]
        segment_ids = [*segment_ids, *[0] * pad_len]

    assert len(input_ids) == max_seq_length
    assert len(input_mask) == max_seq_length
    assert len(segment_ids) == max_seq_length

    label_id = label_map[example.label]
    if ex_index < 5:
        tf.logging.info(">>>>> Example >>>>>")
        tf.logging.info("guid: %s" % (example.guid))
        tf.logging.info("tokens: %s" % " ".join(
            [tokenization.printable_text(x) for x in tokens]))
        tf.logging.info("input_ids: %s" %
                        " ".join([str(x) for x in input_ids]))
        tf.logging.info("input_mask: %s" %
                        " ".join([str(x) for x in input_mask]))
        tf.logging.info("segment_ids: %s" %
                        " ".join([str(x) for x in segment_ids]))
        tf.logging.info("label: %s (id = %d)" % (example.label, label_id))

    feature = InputFeatures(
        input_ids=input_ids,
        input_mask=input_mask,
        segment_ids=segment_ids,
        label_id=label_id,
        is_real_example=True)
    return feature


def file_based_convert_examples_to_features(examples, label_list, max_seq_length, tokenizer, output_file):
    ''' Convert a set of `InputClass`s to a TFRecord file. '''

    writer = tf.python_io.TFRecordWriter(output_file)

    for (ex_index, example) in enumerate(examples):
        if not ex_index % 10000:
            tf.logging.info("Writing example %d of %d" %
                            (ex_index, len(examples)))

        feature = convert_single_example(ex_index, example, label_list,
                                         max_seq_length, tokenizer)

        def create_int_feature(values):
            return tf.train.Feature(
                int64_list=tf.train.Int64List(value=list(values)))

        features = collections.OrderedDict()
        features["input_ids"] = create_int_feature(feature.input_ids)
        features["input_mask"] = create_int_feature(feature.input_mask)
        features["segment_ids"] = create_int_feature(feature.segment_ids)
        features["label_ids"] = create_int_feature([feature.label_id])
        features["is_real_example"] = create_int_feature(
            [int(feature.is_real_example)])

        tf_example = tf.train.Example(
            features=tf.train.Features(feature=features))
        writer.write(tf_example.SerializeToString())
    writer.close()


def file_based_input_fn_builder(input_file, seq_length, is_training, drop_remainder):
    ''' Creates an `input_fn` closure to be passed to TPUEstimator. '''

    name_to_features = {
        "input_ids": tf.FixedLenFeature([seq_length], tf.int64),
        "input_mask": tf.FixedLenFeature([seq_length], tf.int64),
        "segment_ids": tf.FixedLenFeature([seq_length], tf.int64),
        "label_ids": tf.FixedLenFeature([], tf.int64),
        "is_real_example": tf.FixedLenFeature([], tf.int64),
    }

    def _decode_record(record, name_to_features):
        """Decodes a record to a TensorFlow example."""
        example = tf.parse_single_example(record, name_to_features)

        # tf.Example only supports tf.int64, but the TPU only supports tf.int32.
        example = {tf.to_int32(jj) if jj.type ==
                   tf.int64 else jj for ii, jj in example.items()}

        return example

    def input_fn(params):
        """The actual input function."""
        batch_size = params["batch_size"]

        # For training, we want a lot of parallel reading and shuffling.
        # For eval, we want no shuffling and parallel reading doesn't matter.
        d = tf.data.TFRecordDataset(input_file)
        if is_training:
            d = d.repeat()
            d = d.shuffle(buffer_size=100)

        d = d.apply(
            tf.contrib.data.map_and_batch(
                lambda record: _decode_record(record, name_to_features),
                batch_size=batch_size,
                drop_remainder=drop_remainder))

        return d

    return input_fn


def _truncate_seq_pair(tokens_a, tokens_b, max_length):
    '''Truncates a sequence pair in place to the maximum length.'''

    while True:
        total_length = len(tokens_a) + len(tokens_b)
        if total_length <= max_length:
            break
        if len(tokens_a) > len(tokens_b):  # truncate one by one
            tokens_a.pop()
        else:
            tokens_b.pop()


data = {
    'sAct': 'KMdd_StructWebAjax|GetPoisByTag',
    'iMddid': 10186,
    'iTagId': 0,
    'iPage': 2,
    '_ts': int(time.time() * 1000),
    '_sn': '22af158009',
}
test = basic_req(url, 11, data)

header = {
    'Accept': 'application/json, text/javascript, */*; q=0.01',
    'Accept-Encoding': 'gzip, deflate',
    'Accept-Language': 'zh-CN,zh;q=0.9',
    'Cache-Control': 'no-cache',
    'Connection': 'keep-alive',
    'Content-Length': '101',
    'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
    'Cookie': 'PHPSESSID=rom081ou52nseihn8v3th42nj6; mfw_uuid=5cb5438f-9953-52ef-8ae0-b0aec7d2b3d2; _r=google; _rp=a%3A2%3A%7Bs%3A1%3A%22p%22%3Bs%3A18%3A%22www.google.com.hk%2F%22%3Bs%3A1%3A%22t%22%3Bi%3A1555383183%3B%7D; oad_n=a%3A5%3A%7Bs%3A5%3A%22refer%22%3Bs%3A25%3A%22https%3A%2F%2Fwww.google.com.hk%22%3Bs%3A2%3A%22hp%22%3Bs%3A17%3A%22www.google.com.hk%22%3Bs%3A3%3A%22oid%22%3Bi%3A1075%3Bs%3A2%3A%22dm%22%3Bs%3A15%3A%22www.mafengwo.cn%22%3Bs%3A2%3A%22ft%22%3Bs%3A19%3A%222019-04-16+10%3A53%3A03%22%3B%7D; __mfwothchid=referrer%7Cwww.google.com.hk; uva=s%3A156%3A%22a%3A4%3A%7Bs%3A13%3A%22host_pre_time%22%3Bs%3A10%3A%222019-04-16%22%3Bs%3A2%3A%22lt%22%3Bi%3A1555383186%3Bs%3A10%3A%22last_refer%22%3Bs%3A26%3A%22https%3A%2F%2Fwww.google.com.hk%2F%22%3Bs%3A5%3A%22rhost%22%3Bs%3A17%3A%22www.google.com.hk%22%3B%7D%22%3B; __mfwurd=a%3A3%3A%7Bs%3A6%3A%22f_time%22%3Bi%3A1555383186%3Bs%3A9%3A%22f_rdomain%22%3Bs%3A17%3A%22www.google.com.hk%22%3Bs%3A6%3A%22f_host%22%3Bs%3A3%3A%22www%22%3B%7D; __mfwuuid=5cb5438f-9953-52ef-8ae0-b0aec7d2b3d2; UM_distinctid=16a240ffd45627-0212ee25ad3c8f-6d330e7a-1aeaa0-16a240ffd46bc9; CNZZDATA30065558=cnzz_eid%3D168562151-1555380791-http%253A%252F%252Fwww.mafengwo.cn%252F%26ntime%3D1555380791; __mfwlv=1555392228; __mfwvn=2; all_ad=1; arp_scroll_position=3043; RT="sl=1&ss=1555392227458&tt=4532&obo=0&sh=1555392231996%3D1%3A0%3A4532&dm=mafengwo.cn&si=c7b64f9a-b3a2-4238-8374-3d41db513e8d&ld=1555392231996&r=http%3A%2F%2Fwww.mafengwo.cn%2Fjd%2F10186%2Fgonglve.html&ul=1555392241178&hd=1555392241894"; __mfwlt=1555392243',
    'Host': 'www.mafengwo.cn',
    'Origin': 'http://www.mafengwo.cn',
    'Pragma': 'no-cache',
    'Referer': 'http://www.mafengwo.cn/jd/10186/gonglve.html',
    'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/75.0.3766.0 Safari/537.36',
    'X-Requested-With': 'XMLHttpRequest',
}


def create_model(bert_config, is_training, input_ids, input_mask, segment_ids, labels, num_labels, use_one_hot_embeddings):
    ''' Creates a classification model. '''
    model = modeling.BertModel(
        config=bert_config,
        is_training=is_training,
        input_ids=input_ids,
        input_mask=input_mask,
        token_type_ids=segment_ids,
        use_one_hot_embeddings=use_one_hot_embeddings)

    # In the demo, we are doing a simple classification task on the entire segment.
    # If you want to use the token-level output, use model.get_sequence_output()
    output_layer = model.get_pooled_output()  # get word embedding

    hidden_size = output_layer.shape[-1].value

    output_weights = tf.get_variable(
        "output_weights", [num_labels, hidden_size],
        initializer=tf.truncated_normal_initializer(stddev=0.02))

    output_bias = tf.get_variable(
        "output_bias", [num_labels], initializer=tf.zeros_initializer())

    with tf.variable_scope("loss"):
        if is_training:
            # I.e., 0.1 dropout
            output_layer = tf.nn.dropout(output_layer, keep_prob=0.9)

        logits = tf.matmul(output_layer, output_weights, transpose_b=True)
        logits = tf.nn.bias_add(logits, output_bias)
        probabilities = tf.nn.softmax(logits, axis=-1)
        log_probs = tf.nn.log_softmax(logits, axis=-1)

        one_hot_labels = tf.one_hot(labels, depth=num_labels, dtype=tf.float32)

        per_example_loss = -tf.reduce_sum(one_hot_labels * log_probs, axis=-1)
        loss = tf.reduce_mean(per_example_loss)

        return (loss, per_example_loss, logits, probabilities)


def model_fn_builder(bert_config, num_labels, init_checkpoint, learning_rate,
                     num_train_steps, num_warmup_steps, use_tpu,
                     use_one_hot_embeddings):
    """Returns `model_fn` closure for TPUEstimator."""

    def model_fn(features, labels, mode, params):  # pylint: disable=unused-argument
        """The `model_fn` for TPUEstimator."""

        tf.logging.info("*** Features ***")
        for name in sorted(features.keys()):
            tf.logging.info("  name = %s, shape = %s" %
                            (name, features[name].shape))

        input_ids = features["input_ids"]
        input_mask = features["input_mask"]
        segment_ids = features["segment_ids"]
        label_ids = features["label_ids"]
        is_real_example = None
        if "is_real_example" in features:
            is_real_example = tf.cast(
                features["is_real_example"], dtype=tf.float32)
        else:
            is_real_example = tf.ones(tf.shape(label_ids), dtype=tf.float32)

        is_training = (mode == tf.estimator.ModeKeys.TRAIN)

        (total_loss, per_example_loss, logits, probabilities) = create_model(
            bert_config, is_training, input_ids, input_mask, segment_ids, label_ids,
            num_labels, use_one_hot_embeddings)

        tvars = tf.trainable_variables()
        initialized_variable_names = {}
        scaffold_fn = None
        if init_checkpoint:
            (assignment_map, initialized_variable_names
             ) = modeling.get_assignment_map_from_checkpoint(tvars, init_checkpoint)
            if use_tpu:

                def tpu_scaffold():
                    tf.train.init_from_checkpoint(
                        init_checkpoint, assignment_map)
                    return tf.train.Scaffold()

                scaffold_fn = tpu_scaffold
            else:
                tf.train.init_from_checkpoint(init_checkpoint, assignment_map)

        tf.logging.info("**** Trainable Variables ****")
        for var in tvars:
            init_string = ""
            if var.name in initialized_variable_names:
                init_string = ", *INIT_FROM_CKPT*"
            tf.logging.info("  name = %s, shape = %s%s", var.name, var.shape,
                            init_string)

        output_spec = None
        if mode == tf.estimator.ModeKeys.TRAIN:

            train_op = optimization.create_optimizer(
                total_loss, learning_rate, num_train_steps, num_warmup_steps, use_tpu)

            output_spec = tf.contrib.tpu.TPUEstimatorSpec(
                mode=mode,
                loss=total_loss,
                train_op=train_op,
                scaffold_fn=scaffold_fn)
        elif mode == tf.estimator.ModeKeys.EVAL:

            def metric_fn(per_example_loss, label_ids, logits, is_real_example):
                predictions = tf.argmax(logits, axis=-1, output_type=tf.int32)
                accuracy = tf.metrics.accuracy(
                    labels=label_ids, predictions=predictions, weights=is_real_example)
                loss = tf.metrics.mean(
                    values=per_example_loss, weights=is_real_example)
                # f1 = tf.contrib.metrics.f1_score(
                #     labels=label_ids, predictions=predictions, weights=is_real_example)
                # r = tf.metrics.recall(
                #     labels=label_ids, predictions=predictions, weights=is_real_example)
                # p = tf.metrics.precision(
                #     labels=label_ids, predictions=predictions, weights=is_real_example)

                return {
                    # 'r': r,
                    # 'p': p,
                    # 'f1': f1,
                    "eval_accuracy": accuracy,
                    "eval_loss": loss,
                }

            eval_metrics = (metric_fn,
                            [per_example_loss, label_ids, logits, is_real_example])
            output_spec = tf.contrib.tpu.TPUEstimatorSpec(
                mode=mode,
                loss=total_loss,
                eval_metrics=eval_metrics,
                scaffold_fn=scaffold_fn)
        else:
            output_spec = tf.contrib.tpu.TPUEstimatorSpec(
                mode=mode,
                predictions={"probabilities": probabilities},
                scaffold_fn=scaffold_fn)
        return output_spec

    return model_fn


# This function is not used by this file but is still used by the Colab and
# people who depend on it.
def input_fn_builder(features, seq_length, is_training, drop_remainder):
    """Creates an `input_fn` closure to be passed to TPUEstimator."""

    all_input_ids = []
    all_input_mask = []
    all_segment_ids = []
    all_label_ids = []

    for feature in features:
        all_input_ids.append(feature.input_ids)
        all_input_mask.append(feature.input_mask)
        all_segment_ids.append(feature.segment_ids)
        all_label_ids.append(feature.label_id)

    def input_fn(params):
        """The actual input function."""
        batch_size = params["batch_size"]

        num_examples = len(features)

        # This is for demo purposes and does NOT scale to large data sets. We do
        # not use Dataset.from_generator() because that uses tf.py_func which is
        # not TPU compatible. The right way to load data is with TFRecordReader.
        d = tf.data.Dataset.from_tensor_slices({
            "input_ids":
                tf.constant(
                    all_input_ids, shape=[num_examples, seq_length],
                    dtype=tf.int32),
            "input_mask":
                tf.constant(
                    all_input_mask,
                    shape=[num_examples, seq_length],
                    dtype=tf.int32),
            "segment_ids":
                tf.constant(
                    all_segment_ids,
                    shape=[num_examples, seq_length],
                    dtype=tf.int32),
            "label_ids":
                tf.constant(all_label_ids, shape=[
                            num_examples], dtype=tf.int32),
        })

        if is_training:
            d = d.repeat()
            d = d.shuffle(buffer_size=100)

        d = d.batch(batch_size=batch_size, drop_remainder=drop_remainder)
        return d

    return input_fn


# This function is not used by this file but is still used by the Colab and
# people who depend on it.
def convert_examples_to_features(examples, label_list, max_seq_length,
                                 tokenizer):
    """Convert a set of `InputClass`s to a list of `InputFeatures`."""

    features = []
    for (ex_index, example) in enumerate(examples):
        if ex_index % 10000 == 0:
            tf.logging.info("Writing example %d of %d" %
                            (ex_index, len(examples)))

        feature = convert_single_example(ex_index, example, label_list,
                                         max_seq_length, tokenizer)

        features.append(feature)
    return features


def main(_):
    tf.logging.set_verbosity(tf.logging.INFO)

    processors = {"semeval2017task4": SemEval2017Task4Processor}

    tokenization.validate_case_matches_checkpoint(FLAGS.do_lower_case,
                                                  FLAGS.init_checkpoint)

    if not FLAGS.do_train and not FLAGS.do_eval and not FLAGS.do_predict:
        raise ValueError(
            "At least one of `do_train`, `do_eval` or `do_predict' must be True.")

    bert_config = modeling.BertConfig.from_json_file(FLAGS.bert_config_file)

    if FLAGS.max_seq_length > bert_config.max_position_embeddings:
        raise ValueError(
            "Cannot use sequence length %d because the BERT model "
            "was only trained up to sequence length %d" %
            (FLAGS.max_seq_length, bert_config.max_position_embeddings))

    tf.gfile.MakeDirs(FLAGS.output_dir)

    task_name = FLAGS.task_name.lower()

    if task_name not in processors:
        raise ValueError("Task not found: %s" % (task_name))

    processor = processors[task_name]()

    label_list = processor.get_labels()

    tokenizer = tokenization.FullTokenizer(
        vocab_file=FLAGS.vocab_file, do_lower_case=FLAGS.do_lower_case)

    tpu_cluster_resolver = None
    if FLAGS.use_tpu and FLAGS.tpu_name:
        tpu_cluster_resolver = tf.contrib.cluster_resolver.TPUClusterResolver(
            FLAGS.tpu_name, zone=FLAGS.tpu_zone, project=FLAGS.gcp_project)

    is_per_host = tf.contrib.tpu.InputPipelineConfig.PER_HOST_V2
    run_config = tf.contrib.tpu.RunConfig(
        cluster=tpu_cluster_resolver,
        master=FLAGS.master,
        model_dir=FLAGS.output_dir,
        save_checkpoints_steps=FLAGS.save_checkpoints_steps,
        tpu_config=tf.contrib.tpu.TPUConfig(
            iterations_per_loop=FLAGS.iterations_per_loop,
            num_shards=FLAGS.num_tpu_cores,
            per_host_input_for_training=is_per_host))

    train_examples = None
    num_train_steps = None
    num_warmup_steps = None
    if FLAGS.do_train:
        train_examples = processor.get_train_examples(FLAGS.data_dir)
        num_train_steps = int(
            len(train_examples) / FLAGS.train_batch_size * FLAGS.num_train_epochs)
        num_warmup_steps = int(num_train_steps * FLAGS.warmup_proportion)

    model_fn = model_fn_builder(
        bert_config=bert_config,
        num_labels=len(label_list),
        init_checkpoint=FLAGS.init_checkpoint,
        learning_rate=FLAGS.learning_rate,
        num_train_steps=num_train_steps,
        num_warmup_steps=num_warmup_steps,
        use_tpu=FLAGS.use_tpu,
        use_one_hot_embeddings=FLAGS.use_tpu)

    # If TPU is not available, this will fall back to normal Estimator on CPU
    # or GPU.
    estimator = tf.contrib.tpu.TPUEstimator(
        use_tpu=FLAGS.use_tpu,
        model_fn=model_fn,
        config=run_config,
        train_batch_size=FLAGS.train_batch_size,
        eval_batch_size=FLAGS.eval_batch_size,
        predict_batch_size=FLAGS.predict_batch_size)

    if FLAGS.do_train:
        train_file = os.path.join(FLAGS.output_dir, "train.tf_record")
        file_based_convert_examples_to_features(
            train_examples, label_list, FLAGS.max_seq_length, tokenizer, train_file)
        tf.logging.info("***** Running training *****")
        tf.logging.info("  Num examples = %d", len(train_examples))
        tf.logging.info("  Batch size = %d", FLAGS.train_batch_size)
        tf.logging.info("  Num steps = %d", num_train_steps)
        train_input_fn = file_based_input_fn_builder(
            input_file=train_file,
            seq_length=FLAGS.max_seq_length,
            is_training=True,
            drop_remainder=True)
        estimator.train(input_fn=train_input_fn, max_steps=num_train_steps)

    if FLAGS.do_eval:
        eval_examples = processor.get_dev_examples(FLAGS.data_dir)
        num_actual_eval_examples = len(eval_examples)
        if FLAGS.use_tpu:
            # TPU requires a fixed batch size for all batches, therefore the number
            # of examples must be a multiple of the batch size, or else examples
            # will get dropped. So we pad with fake examples which are ignored
            # later on. These do NOT count towards the metric (all tf.metrics
            # support a per-instance weight, and these get a weight of 0.0).
            while len(eval_examples) % FLAGS.eval_batch_size != 0:
                eval_examples.append(PaddingClass())

        eval_file = os.path.join(FLAGS.output_dir, "eval.tf_record")
        file_based_convert_examples_to_features(
            eval_examples, label_list, FLAGS.max_seq_length, tokenizer, eval_file)

        tf.logging.info("***** Running evaluation *****")
        tf.logging.info("  Num examples = %d (%d actual, %d padding)",
                        len(eval_examples), num_actual_eval_examples,
                        len(eval_examples) - num_actual_eval_examples)
        tf.logging.info("  Batch size = %d", FLAGS.eval_batch_size)

        # This tells the estimator to run through the entire set.
        eval_steps = None
        # However, if running eval on the TPU, you will need to specify the
        # number of steps.
        if FLAGS.use_tpu:
            assert len(eval_examples) % FLAGS.eval_batch_size == 0
            eval_steps = int(len(eval_examples) // FLAGS.eval_batch_size)

        eval_drop_remainder = True if FLAGS.use_tpu else False
        eval_input_fn = file_based_input_fn_builder(
            input_file=eval_file,
            seq_length=FLAGS.max_seq_length,
            is_training=False,
            drop_remainder=eval_drop_remainder)

        result = estimator.evaluate(input_fn=eval_input_fn, steps=eval_steps)

        output_eval_file = os.path.join(FLAGS.output_dir, "eval_results.txt")
        with tf.gfile.GFile(output_eval_file, "w") as writer:
            tf.logging.info("***** Eval results *****")
            for key in sorted(result.keys()):
                tf.logging.info("  %s = %s", key, str(result[key]))
                writer.write("%s = %s\n" % (key, str(result[key])))

    if FLAGS.do_predict:
        predict_examples = processor.get_test_examples(FLAGS.data_dir)
        num_actual_predict_examples = len(predict_examples)
        if FLAGS.use_tpu:
            # TPU requires a fixed batch size for all batches, therefore the number
            # of examples must be a multiple of the batch size, or else examples
            # will get dropped. So we pad with fake examples which are ignored
            # later on.
            while len(predict_examples) % FLAGS.predict_batch_size != 0:
                predict_examples.append(PaddingClass())

        predict_file = os.path.join(FLAGS.output_dir, "predict.tf_record")
        file_based_convert_examples_to_features(predict_examples, label_list,
                                                FLAGS.max_seq_length, tokenizer,
                                                predict_file)

        tf.logging.info("***** Running prediction*****")
        tf.logging.info("  Num examples = %d (%d actual, %d padding)",
                        len(predict_examples), num_actual_predict_examples,
                        len(predict_examples) - num_actual_predict_examples)
        tf.logging.info("  Batch size = %d", FLAGS.predict_batch_size)

        predict_drop_remainder = True if FLAGS.use_tpu else False
        predict_input_fn = file_based_input_fn_builder(
            input_file=predict_file,
            seq_length=FLAGS.max_seq_length,
            is_training=False,
            drop_remainder=predict_drop_remainder)

        result = estimator.predict(input_fn=predict_input_fn)

        output_predict_file = os.path.join(
            FLAGS.output_dir, "test_results_{}_{}.tsv".format(FLAGS.pad_type, FLAGS.ekphrasis))
        with tf.gfile.GFile(output_predict_file, "w") as writer:
            num_written_lines = 0
            tf.logging.info("***** Predict results *****")
            for (i, prediction) in enumerate(result):
                probabilities = prediction["probabilities"]
                if i >= num_actual_predict_examples:
                    break
                output_line = "\t".join(
                    str(class_probability)
                    for class_probability in probabilities) + "\n"
                writer.write(output_line)
                num_written_lines += 1
        assert num_written_lines == num_actual_predict_examples
        predict = [np.argmax(np.array(ii)) for ii in result]
        # label = processor.text_Y
        # p, r, f1, _, _, acc = load_result_f1(predict, label)


if __name__ == "__main__":
    flags.mark_flag_as_required("data_dir")
    flags.mark_flag_as_required("task_name")
    flags.mark_flag_as_required("vocab_file")
    flags.mark_flag_as_required("bert_config_file")
    # flags.mark_flag_as_required("output_dir")
    tf.app.run()