Merge pull request #6 from tensorflow/master

Updated

official/README.md

@@ -80,6 +80,8 @@ installable Official Models package. This is being tracked in
 
 ### Natural Language Processing
 
+*   [albert](nlp/albert): A Lite BERT for Self-supervised Learning of Language
+    Representations.
 *   [bert](nlp/bert): A powerful pre-trained language representation model:
     BERT, which stands for Bidirectional Encoder Representations from
     Transformers.

official/benchmark/bert_benchmark.py

@@ -212,39 +212,6 @@ class BertClassifyBenchmarkReal(BertClassifyBenchmarkBase):
                                 'summaries/training_summary.txt')
     self._run_and_report_benchmark(summary_path, use_ds=False)
 
-  def benchmark_2_gpu_mrpc(self):
-    """Test BERT model performance with 2 GPUs."""
-
-    self._setup()
-    self.num_gpus = 2
-    FLAGS.model_dir = self._get_model_dir('benchmark_2_gpu_mrpc')
-    FLAGS.train_data_path = self.train_data_path
-    FLAGS.eval_data_path = self.eval_data_path
-    FLAGS.input_meta_data_path = self.input_meta_data_path
-    FLAGS.bert_config_file = self.bert_config_file
-    FLAGS.train_batch_size = 8
-    FLAGS.eval_batch_size = 8
-
-    summary_path = os.path.join(FLAGS.model_dir,
-                                'summaries/training_summary.txt')
-    self._run_and_report_benchmark(summary_path)
-
-  def benchmark_4_gpu_mrpc(self):
-    """Test BERT model performance with 4 GPUs."""
-
-    self._setup()
-    self.num_gpus = 4
-    FLAGS.model_dir = self._get_model_dir('benchmark_4_gpu_mrpc')
-    FLAGS.train_data_path = self.train_data_path
-    FLAGS.eval_data_path = self.eval_data_path
-    FLAGS.input_meta_data_path = self.input_meta_data_path
-    FLAGS.bert_config_file = self.bert_config_file
-    FLAGS.train_batch_size = 16
-
-    summary_path = os.path.join(FLAGS.model_dir,
-                                'summaries/training_summary.txt')
-    self._run_and_report_benchmark(summary_path)
-
   def benchmark_8_gpu_mrpc(self):
     """Test BERT model performance with 8 GPUs."""
 

official/benchmark/bert_squad_benchmark.py

@@ -24,12 +24,12 @@ import time
 
 # pylint: disable=g-bad-import-order
 from absl import flags
+from absl import logging
 from absl.testing import flagsaver
 import tensorflow as tf
 # pylint: enable=g-bad-import-order
 
 from official.benchmark import bert_benchmark_utils as benchmark_utils
-from official.benchmark import squad_evaluate_v1_1
 from official.nlp.bert import run_squad
 from official.utils.misc import distribution_utils
 from official.utils.misc import keras_utils
@@ -70,18 +70,6 @@ class BertSquadBenchmarkBase(benchmark_utils.BertBenchmarkBase):
     with tf.io.gfile.GFile(FLAGS.input_meta_data_path, 'rb') as reader:
       return json.loads(reader.read().decode('utf-8'))
 
-  def _read_predictions_dataset_from_file(self):
-    """Reads the predictions dataset from a file."""
-    with tf.io.gfile.GFile(SQUAD_PREDICT_FILE, 'r') as reader:
-      dataset_json = json.load(reader)
-      return dataset_json['data']
-
-  def _read_predictions_from_file(self):
-    """Reads the predictions from a file."""
-    predictions_file = os.path.join(FLAGS.model_dir, 'predictions.json')
-    with tf.io.gfile.GFile(predictions_file, 'r') as reader:
-      return json.load(reader)
-
   def _get_distribution_strategy(self, ds_type='mirrored'):
     """Gets the distribution strategy.
 
@@ -135,12 +123,10 @@ class BertSquadBenchmarkBase(benchmark_utils.BertBenchmarkBase):
     input_meta_data = self._read_input_meta_data_from_file()
     strategy = self._get_distribution_strategy(ds_type)
 
-    run_squad.predict_squad(strategy=strategy, input_meta_data=input_meta_data)
-
-    dataset = self._read_predictions_dataset_from_file()
-    predictions = self._read_predictions_from_file()
-
-    eval_metrics = squad_evaluate_v1_1.evaluate(dataset, predictions)
+    if input_meta_data.get('version_2_with_negative', False):
+      logging.error('In memory evaluation result for SQuAD v2 is not accurate')
+    eval_metrics = run_squad.eval_squad(strategy=strategy,
+                                        input_meta_data=input_meta_data)
     # Use F1 score as reported evaluation metric.
     self.eval_metrics = eval_metrics['f1']
 

official/benchmark/models/resnet_cifar_main.py

@@ -20,6 +20,7 @@ from __future__ import print_function
 
 from absl import app
 from absl import flags
+from absl import logging
 import numpy as np
 import tensorflow as tf
 from official.benchmark.models import resnet_cifar_model
@@ -100,7 +101,7 @@ class LearningRateBatchScheduler(tf.keras.callbacks.Callback):
     if lr != self.prev_lr:
       self.model.optimizer.learning_rate = lr  # lr should be a float here
       self.prev_lr = lr
-      tf.compat.v1.logging.debug(
+      logging.debug(
           'Epoch %05d Batch %05d: LearningRateBatchScheduler '
           'change learning rate to %s.', self.epochs, batch, lr)
 
@@ -137,8 +138,8 @@ def run(flags_obj):
 
   data_format = flags_obj.data_format
   if data_format is None:
-    data_format = ('channels_first'
-                   if tf.test.is_built_with_cuda() else 'channels_last')
+    data_format = ('channels_first' if tf.config.list_physical_devices('GPU')
+                   else 'channels_last')
   tf.keras.backend.set_image_data_format(data_format)
 
   strategy = distribution_utils.get_distribution_strategy(
@@ -280,6 +281,6 @@ def main(_):
 
 
 if __name__ == '__main__':
-  tf.compat.v1.logging.set_verbosity(tf.compat.v1.logging.INFO)
+  logging.set_verbosity(logging.INFO)
   define_cifar_flags()
   app.run(main)

official/benchmark/ncf_keras_benchmark.py

@@ -13,7 +13,6 @@
 # limitations under the License.
 # ==============================================================================
 """Executes Keras benchmarks and accuracy tests."""
-
 from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
@@ -22,6 +21,7 @@ import os
 import time
 
 from absl import flags
+from absl import logging
 from absl.testing import flagsaver
 import tensorflow as tf
 
@@ -51,7 +51,7 @@ class NCFKerasBenchmarkBase(tf.test.Benchmark):
   def _setup(self):
     """Sets up and resets flags before each test."""
     assert tf.version.VERSION.startswith('2.')
-    tf.compat.v1.logging.set_verbosity(tf.compat.v1.logging.INFO)
+    logging.set_verbosity(logging.INFO)
     if NCFKerasBenchmarkBase.local_flags is None:
       ncf_common.define_ncf_flags()
       # Loads flags to get defaults to then override. List cannot be empty.

official/nlp/bert/README.md

@@ -269,6 +269,7 @@ python run_classifier.py \
   --init_checkpoint=${BERT_DIR}/bert_model.ckpt \
   --train_batch_size=32 \
   --eval_batch_size=32 \
+  --steps_per_loop=1000 \
   --learning_rate=2e-5 \
   --num_train_epochs=3 \
   --model_dir=${MODEL_DIR} \
@@ -276,6 +277,10 @@ python run_classifier.py \
   --tpu=grpc://${TPU_IP_ADDRESS}:8470
 ```
 
+Note that, we specify `steps_per_loop=1000` for TPU, because running a loop of
+training steps inside a `tf.function` can significantly increase TPU utilization
+and callbacks will not be called inside the loop.
+
 ### SQuAD 1.1
 
 The Stanford Question Answering Dataset (SQuAD) is a popular question answering

official/nlp/bert/common_flags.py

@@ -57,7 +57,7 @@ def define_common_bert_flags():
   flags.DEFINE_integer('num_train_epochs', 3,
                        'Total number of training epochs to perform.')
   flags.DEFINE_integer(
-      'steps_per_loop', 200,
+      'steps_per_loop', 1,
       'Number of steps per graph-mode loop. Only training step '
       'happens inside the loop. Callbacks will not be called '
       'inside.')

official/modeling/model_training_utils.py → official/nlp/bert/model_training_utils.py

@@ -415,7 +415,7 @@ def run_customized_training_loop(
       # Runs several steps in the host while loop.
       steps = steps_to_run(current_step, steps_per_epoch, steps_per_loop)
 
-      if tf.test.is_built_with_cuda():
+      if tf.config.list_physical_devices('GPU'):
         # TODO(zongweiz): merge with train_steps once tf.while_loop
         # GPU performance bugs are fixed.
         for _ in range(steps):

official/modeling/model_training_utils_test.py → official/nlp/bert/model_training_utils_test.py

@@ -20,6 +20,7 @@ from __future__ import print_function
 
 import os
 
+from absl import logging
 from absl.testing import parameterized
 from absl.testing.absltest import mock
 import numpy as np
@@ -27,7 +28,7 @@ import tensorflow as tf
 
 from tensorflow.python.distribute import combinations
 from tensorflow.python.distribute import strategy_combinations
-from official.modeling import model_training_utils
+from official.nlp.bert import model_training_utils
 
 
 def eager_strategy_combinations():
@@ -125,7 +126,7 @@ def summaries_with_matching_keyword(keyword, summary_dir):
     if event.summary is not None:
       for value in event.summary.value:
         if keyword in value.tag:
-          tf.compat.v1.logging.error(event)
+          logging.error(event)
           yield event.summary
 
 

official/nlp/bert/run_classifier.py

@@ -25,8 +25,6 @@ from absl import app
 from absl import flags
 from absl import logging
 import tensorflow as tf
-
-from official.modeling import model_training_utils
 from official.modeling import performance
 from official.nlp import optimization
 from official.nlp.bert import bert_models
@@ -34,6 +32,7 @@ from official.nlp.bert import common_flags
 from official.nlp.bert import configs as bert_configs
 from official.nlp.bert import input_pipeline
 from official.nlp.bert import model_saving_utils
+from official.nlp.bert import model_training_utils
 from official.utils.misc import distribution_utils
 from official.utils.misc import keras_utils
 
@@ -156,6 +155,7 @@ def run_bert_classifier(strategy,
         init_checkpoint,
         epochs,
         steps_per_epoch,
+        steps_per_loop,
         eval_steps,
         custom_callbacks=custom_callbacks)
 
@@ -189,6 +189,7 @@ def run_keras_compile_fit(model_dir,
                           init_checkpoint,
                           epochs,
                           steps_per_epoch,
+                          steps_per_loop,
                           eval_steps,
                           custom_callbacks=None):
   """Runs BERT classifier model using Keras compile/fit API."""
@@ -203,7 +204,11 @@ def run_keras_compile_fit(model_dir,
       checkpoint = tf.train.Checkpoint(model=sub_model)
       checkpoint.restore(init_checkpoint).assert_existing_objects_matched()
 
-    bert_model.compile(optimizer=optimizer, loss=loss_fn, metrics=[metric_fn()])
+    bert_model.compile(
+        optimizer=optimizer,
+        loss=loss_fn,
+        metrics=[metric_fn()],
+        experimental_steps_per_execution=steps_per_loop)
 
     summary_dir = os.path.join(model_dir, 'summaries')
     summary_callback = tf.keras.callbacks.TensorBoard(summary_dir)

official/nlp/bert/run_pretraining.py

@@ -22,14 +22,13 @@ from absl import flags
 from absl import logging
 import gin
 import tensorflow as tf
-
-from official.modeling import model_training_utils
 from official.modeling import performance
 from official.nlp import optimization
 from official.nlp.bert import bert_models
 from official.nlp.bert import common_flags
 from official.nlp.bert import configs
 from official.nlp.bert import input_pipeline
+from official.nlp.bert import model_training_utils
 from official.utils.misc import distribution_utils
 
 

official/nlp/bert/run_squad.py

@@ -19,9 +19,13 @@ from __future__ import division
 from __future__ import print_function
 
 import json
+import os
+import tempfile
+import time
 
 from absl import app
 from absl import flags
+from absl import logging
 import tensorflow as tf
 
 from official.nlp.bert import configs as bert_configs
@@ -52,12 +56,22 @@ def train_squad(strategy,
 
 
 def predict_squad(strategy, input_meta_data):
-  """Makes predictions for a squad dataset."""
+  """Makes predictions for the squad dataset."""
   bert_config = bert_configs.BertConfig.from_json_file(FLAGS.bert_config_file)
   tokenizer = tokenization.FullTokenizer(
       vocab_file=FLAGS.vocab_file, do_lower_case=FLAGS.do_lower_case)
-  run_squad_helper.predict_squad(strategy, input_meta_data, tokenizer,
-                                 bert_config, squad_lib_wp)
+  run_squad_helper.predict_squad(
+      strategy, input_meta_data, tokenizer, bert_config, squad_lib_wp)
+
+
+def eval_squad(strategy, input_meta_data):
+  """Evaluate on the squad dataset."""
+  bert_config = bert_configs.BertConfig.from_json_file(FLAGS.bert_config_file)
+  tokenizer = tokenization.FullTokenizer(
+      vocab_file=FLAGS.vocab_file, do_lower_case=FLAGS.do_lower_case)
+  eval_metrics = run_squad_helper.eval_squad(
+      strategy, input_meta_data, tokenizer, bert_config, squad_lib_wp)
+  return eval_metrics
 
 
 def export_squad(model_export_path, input_meta_data):
@@ -93,7 +107,8 @@ def main(_):
       num_gpus=FLAGS.num_gpus,
       all_reduce_alg=FLAGS.all_reduce_alg,
       tpu_address=FLAGS.tpu)
-  if FLAGS.mode in ('train', 'train_and_predict'):
+
+  if 'train' in FLAGS.mode:
     if FLAGS.log_steps:
       custom_callbacks = [keras_utils.TimeHistory(
           batch_size=FLAGS.train_batch_size,
@@ -109,8 +124,25 @@ def main(_):
         custom_callbacks=custom_callbacks,
         run_eagerly=FLAGS.run_eagerly,
     )
-  if FLAGS.mode in ('predict', 'train_and_predict'):
+  if 'predict' in FLAGS.mode:
     predict_squad(strategy, input_meta_data)
+  if 'eval' in FLAGS.mode:
+    eval_metrics = eval_squad(strategy, input_meta_data)
+    f1_score = eval_metrics['final_f1']
+    logging.info('SQuAD eval F1-score: %f', f1_score)
+    if (not strategy) or strategy.extended.should_save_summary:
+      summary_dir = os.path.join(FLAGS.model_dir, 'summaries')
+    else:
+      summary_dir = tempfile.mkdtemp()
+    summary_writer = tf.summary.create_file_writer(
+        os.path.join(summary_dir, 'eval'))
+    with summary_writer.as_default():
+      # TODO(lehou): write to the correct step number.
+      tf.summary.scalar('F1-score', f1_score, step=0)
+      summary_writer.flush()
+    # Wait for some time, for the depending mldash/tensorboard jobs to finish
+    # exporting the final F1-score.
+    time.sleep(60)
 
 
 if __name__ == '__main__':

official/nlp/bert/run_squad_helper.py

@@ -18,18 +18,20 @@ from __future__ import division
 from __future__ import print_function
 
 import collections
+import json
 import os
 from absl import flags
 from absl import logging
 import tensorflow as tf
-
-from official.modeling import model_training_utils
 from official.modeling import performance
 from official.nlp import optimization
 from official.nlp.bert import bert_models
 from official.nlp.bert import common_flags
 from official.nlp.bert import input_pipeline
 from official.nlp.bert import model_saving_utils
+from official.nlp.bert import model_training_utils
+from official.nlp.bert import squad_evaluate_v1_1
+from official.nlp.bert import squad_evaluate_v2_0
 from official.nlp.data import squad_lib_sp
 from official.utils.misc import keras_utils
 
@@ -37,11 +39,15 @@ from official.utils.misc import keras_utils
 def define_common_squad_flags():
   """Defines common flags used by SQuAD tasks."""
   flags.DEFINE_enum(
-      'mode', 'train_and_predict',
-      ['train_and_predict', 'train', 'predict', 'export_only'],
-      'One of {"train_and_predict", "train", "predict", "export_only"}. '
-      '`train_and_predict`: both train and predict to a json file. '
+      'mode', 'train_and_eval',
+      ['train_and_eval', 'train_and_predict',
+       'train', 'eval', 'predict', 'export_only'],
+      'One of {"train_and_eval", "train_and_predict", '
+      '"train", "eval", "predict", "export_only"}. '
+      '`train_and_eval`: train & predict to json files & compute eval metrics. '
+      '`train_and_predict`: train & predict to json files. '
       '`train`: only trains the model. '
+      '`eval`: predict answers from squad json file & compute eval metrics. '
       '`predict`: predict answers from the squad json file. '
       '`export_only`: will take the latest checkpoint inside '
       'model_dir and export a `SavedModel`.')
@@ -271,7 +277,8 @@ def train_squad(strategy,
       post_allreduce_callbacks=[clip_by_global_norm_callback])
 
 
-def predict_squad(strategy, input_meta_data, tokenizer, bert_config, squad_lib):
+def prediction_output_squad(
+    strategy, input_meta_data, tokenizer, bert_config, squad_lib):
   """Makes predictions for a squad dataset."""
   doc_stride = input_meta_data['doc_stride']
   max_query_length = input_meta_data['max_query_length']
@@ -322,23 +329,61 @@ def predict_squad(strategy, input_meta_data, tokenizer, bert_config, squad_lib):
   all_results = predict_squad_customized(strategy, input_meta_data, bert_config,
                                          eval_writer.filename, num_steps)
 
-  output_prediction_file = os.path.join(FLAGS.model_dir, 'predictions.json')
-  output_nbest_file = os.path.join(FLAGS.model_dir, 'nbest_predictions.json')
-  output_null_log_odds_file = os.path.join(FLAGS.model_dir, 'null_odds.json')
-
-  squad_lib.write_predictions(
+  all_predictions, all_nbest_json, scores_diff_json = (
+      squad_lib.postprocess_output(
           eval_examples,
           eval_features,
           all_results,
           FLAGS.n_best_size,
           FLAGS.max_answer_length,
           FLAGS.do_lower_case,
-      output_prediction_file,
-      output_nbest_file,
-      output_null_log_odds_file,
           version_2_with_negative=version_2_with_negative,
           null_score_diff_threshold=FLAGS.null_score_diff_threshold,
-      verbose=FLAGS.verbose_logging)
+          verbose=FLAGS.verbose_logging))
+
+  return all_predictions, all_nbest_json, scores_diff_json
+
+
+def dump_to_files(all_predictions, all_nbest_json, scores_diff_json,
+                  squad_lib, version_2_with_negative):
+  """Save output to json files."""
+  output_prediction_file = os.path.join(FLAGS.model_dir, 'predictions.json')
+  output_nbest_file = os.path.join(FLAGS.model_dir, 'nbest_predictions.json')
+  output_null_log_odds_file = os.path.join(FLAGS.model_dir, 'null_odds.json')
+  logging.info('Writing predictions to: %s', (output_prediction_file))
+  logging.info('Writing nbest to: %s', (output_nbest_file))
+
+  squad_lib.write_to_json_files(all_predictions, output_prediction_file)
+  squad_lib.write_to_json_files(all_nbest_json, output_nbest_file)
+  if version_2_with_negative:
+    squad_lib.write_to_json_files(scores_diff_json, output_null_log_odds_file)
+
+
+def predict_squad(strategy, input_meta_data, tokenizer, bert_config, squad_lib):
+  """Get prediction results and evaluate them to hard drive."""
+  all_predictions, all_nbest_json, scores_diff_json = prediction_output_squad(
+      strategy, input_meta_data, tokenizer, bert_config, squad_lib)
+  dump_to_files(all_predictions, all_nbest_json, scores_diff_json, squad_lib,
+                input_meta_data.get('version_2_with_negative', False))
+
+
+def eval_squad(strategy, input_meta_data, tokenizer, bert_config, squad_lib):
+  """Get prediction results and evaluate them against ground truth."""
+  all_predictions, all_nbest_json, scores_diff_json = prediction_output_squad(
+      strategy, input_meta_data, tokenizer, bert_config, squad_lib)
+  dump_to_files(all_predictions, all_nbest_json, scores_diff_json, squad_lib,
+                input_meta_data.get('version_2_with_negative', False))
+
+  with tf.io.gfile.GFile(FLAGS.predict_file, 'r') as reader:
+    dataset_json = json.load(reader)
+    pred_dataset = dataset_json['data']
+  if input_meta_data.get('version_2_with_negative', False):
+    eval_metrics = squad_evaluate_v2_0.evaluate(pred_dataset,
+                                                all_predictions,
+                                                scores_diff_json)
+  else:
+    eval_metrics = squad_evaluate_v1_1.evaluate(pred_dataset, all_predictions)
+  return eval_metrics
 
 
 def export_squad(model_export_path, input_meta_data, bert_config):

official/nlp/bert/squad_evaluate_v1_1.py

+# Copyright 2019 The TensorFlow Authors. All Rights Reserved.
+# 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.
+# ==============================================================================
+"""Evaluation of SQuAD predictions (version 1.1).
+
+The functions are copied from
+https://worksheets.codalab.org/rest/bundles/0xbcd57bee090b421c982906709c8c27e1/contents/blob/.
+
+The SQuAD dataset is described in this paper:
+SQuAD: 100,000+ Questions for Machine Comprehension of Text
+Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev, Percy Liang
+https://nlp.stanford.edu/pubs/rajpurkar2016squad.pdf
+"""
+
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import collections
+import re
+import string
+
+# pylint: disable=g-bad-import-order
+from absl import logging
+# pylint: enable=g-bad-import-order
+
+
+def _normalize_answer(s):
+  """Lowers text and remove punctuation, articles and extra whitespace."""
+
+  def remove_articles(text):
+    return re.sub(r"\b(a|an|the)\b", " ", text)
+
+  def white_space_fix(text):
+    return " ".join(text.split())
+
+  def remove_punc(text):
+    exclude = set(string.punctuation)
+    return "".join(ch for ch in text if ch not in exclude)
+
+  def lower(text):
+    return text.lower()
+
+  return white_space_fix(remove_articles(remove_punc(lower(s))))
+
+
+def _f1_score(prediction, ground_truth):
+  """Computes F1 score by comparing prediction to ground truth."""
+  prediction_tokens = _normalize_answer(prediction).split()
+  ground_truth_tokens = _normalize_answer(ground_truth).split()
+  prediction_counter = collections.Counter(prediction_tokens)
+  ground_truth_counter = collections.Counter(ground_truth_tokens)
+  common = prediction_counter & ground_truth_counter
+  num_same = sum(common.values())
+  if num_same == 0:
+    return 0
+  precision = 1.0 * num_same / len(prediction_tokens)
+  recall = 1.0 * num_same / len(ground_truth_tokens)
+  f1 = (2 * precision * recall) / (precision + recall)
+  return f1
+
+
+def _exact_match_score(prediction, ground_truth):
+  """Checks if predicted answer exactly matches ground truth answer."""
+  return _normalize_answer(prediction) == _normalize_answer(ground_truth)
+
+
+def _metric_max_over_ground_truths(metric_fn, prediction, ground_truths):
+  """Computes the max over all metric scores."""
+  scores_for_ground_truths = []
+  for ground_truth in ground_truths:
+    score = metric_fn(prediction, ground_truth)
+    scores_for_ground_truths.append(score)
+  return max(scores_for_ground_truths)
+
+
+def evaluate(dataset, predictions):
+  """Evaluates predictions for a dataset."""
+  f1 = exact_match = total = 0
+  for article in dataset:
+    for paragraph in article["paragraphs"]:
+      for qa in paragraph["qas"]:
+        total += 1
+        if qa["id"] not in predictions:
+          message = "Unanswered question " + qa["id"] + " will receive score 0."
+          logging.error(message)
+          continue
+        ground_truths = [entry["text"] for entry in qa["answers"]]
+        prediction = predictions[qa["id"]]
+        exact_match += _metric_max_over_ground_truths(_exact_match_score,
+                                                      prediction, ground_truths)
+        f1 += _metric_max_over_ground_truths(_f1_score, prediction,
+                                             ground_truths)
+
+  exact_match = exact_match / total
+  f1 = f1 / total
+
+  return {"exact_match": exact_match, "final_f1": f1}

official/nlp/bert/squad_evaluate_v2_0.py

+# Copyright 2019 The TensorFlow Authors. All Rights Reserved.
+# 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.
+# ==============================================================================
+"""Evaluation script for SQuAD version 2.0.
+
+The functions are copied and modified from
+https://raw.githubusercontent.com/white127/SQUAD-2.0-bidaf/master/evaluate-v2.0.py
+
+In addition to basic functionality, we also compute additional statistics and
+plot precision-recall curves if an additional na_prob.json file is provided.
+This file is expected to map question ID's to the model's predicted probability
+that a question is unanswerable.
+"""
+
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import collections
+import re
+import string
+
+from absl import logging
+
+
+def _make_qid_to_has_ans(dataset):
+  qid_to_has_ans = {}
+  for article in dataset:
+    for p in article['paragraphs']:
+      for qa in p['qas']:
+        qid_to_has_ans[qa['id']] = bool(qa['answers'])
+  return qid_to_has_ans
+
+
+def _normalize_answer(s):
+  """Lower text and remove punctuation, articles and extra whitespace."""
+  def remove_articles(text):
+    regex = re.compile(r'\b(a|an|the)\b', re.UNICODE)
+    return re.sub(regex, ' ', text)
+  def white_space_fix(text):
+    return ' '.join(text.split())
+  def remove_punc(text):
+    exclude = set(string.punctuation)
+    return ''.join(ch for ch in text if ch not in exclude)
+  def lower(text):
+    return text.lower()
+  return white_space_fix(remove_articles(remove_punc(lower(s))))
+
+
+def _get_tokens(s):
+  if not s: return []
+  return _normalize_answer(s).split()
+
+
+def _compute_exact(a_gold, a_pred):
+  return int(_normalize_answer(a_gold) == _normalize_answer(a_pred))
+
+
+def _compute_f1(a_gold, a_pred):
+  """Compute F1-score."""
+  gold_toks = _get_tokens(a_gold)
+  pred_toks = _get_tokens(a_pred)
+  common = collections.Counter(gold_toks) & collections.Counter(pred_toks)
+  num_same = sum(common.values())
+  if not gold_toks or not pred_toks:
+    # If either is no-answer, then F1 is 1 if they agree, 0 otherwise
+    return int(gold_toks == pred_toks)
+  if num_same == 0:
+    return 0
+  precision = 1.0 * num_same / len(pred_toks)
+  recall = 1.0 * num_same / len(gold_toks)
+  f1 = (2 * precision * recall) / (precision + recall)
+  return f1
+
+
+def _get_raw_scores(dataset, predictions):
+  """Compute raw scores."""
+  exact_scores = {}
+  f1_scores = {}
+  for article in dataset:
+    for p in article['paragraphs']:
+      for qa in p['qas']:
+        qid = qa['id']
+        gold_answers = [a['text'] for a in qa['answers']
+                        if _normalize_answer(a['text'])]
+        if not gold_answers:
+          # For unanswerable questions, only correct answer is empty string
+          gold_answers = ['']
+        if qid not in predictions:
+          logging.error('Missing prediction for %s', qid)
+          continue
+        a_pred = predictions[qid]
+        # Take max over all gold answers
+        exact_scores[qid] = max(_compute_exact(a, a_pred) for a in gold_answers)
+        f1_scores[qid] = max(_compute_f1(a, a_pred) for a in gold_answers)
+  return exact_scores, f1_scores
+
+
+def _apply_no_ans_threshold(
+    scores, na_probs, qid_to_has_ans, na_prob_thresh=1.0):
+  new_scores = {}
+  for qid, s in scores.items():
+    pred_na = na_probs[qid] > na_prob_thresh
+    if pred_na:
+      new_scores[qid] = float(not qid_to_has_ans[qid])
+    else:
+      new_scores[qid] = s
+  return new_scores
+
+
+def _make_eval_dict(exact_scores, f1_scores, qid_list=None):
+  """Make evaluation result dictionary."""
+  if not qid_list:
+    total = len(exact_scores)
+    return collections.OrderedDict([
+        ('exact', 100.0 * sum(exact_scores.values()) / total),
+        ('f1', 100.0 * sum(f1_scores.values()) / total),
+        ('total', total),
+    ])
+  else:
+    total = len(qid_list)
+    return collections.OrderedDict([
+        ('exact', 100.0 * sum(exact_scores[k] for k in qid_list) / total),
+        ('f1', 100.0 * sum(f1_scores[k] for k in qid_list) / total),
+        ('total', total),
+    ])
+
+
+def _merge_eval(main_eval, new_eval, prefix):
+  for k in new_eval:
+    main_eval['%s_%s' % (prefix, k)] = new_eval[k]
+
+
+def _make_precision_recall_eval(scores, na_probs, num_true_pos, qid_to_has_ans):
+  """Make evaluation dictionary containing average recision recall."""
+  qid_list = sorted(na_probs, key=lambda k: na_probs[k])
+  true_pos = 0.0
+  cur_p = 1.0
+  cur_r = 0.0
+  precisions = [1.0]
+  recalls = [0.0]
+  avg_prec = 0.0
+  for i, qid in enumerate(qid_list):
+    if qid_to_has_ans[qid]:
+      true_pos += scores[qid]
+    cur_p = true_pos / float(i+1)
+    cur_r = true_pos / float(num_true_pos)
+    if i == len(qid_list) - 1 or na_probs[qid] != na_probs[qid_list[i+1]]:
+      # i.e., if we can put a threshold after this point
+      avg_prec += cur_p * (cur_r - recalls[-1])
+      precisions.append(cur_p)
+      recalls.append(cur_r)
+  return {'ap': 100.0 * avg_prec}
+
+
+def _run_precision_recall_analysis(
+    main_eval, exact_raw, f1_raw, na_probs, qid_to_has_ans):
+  """Run precision recall analysis and return result dictionary."""
+  num_true_pos = sum(1 for v in qid_to_has_ans.values() if v)
+  if num_true_pos == 0:
+    return
+  pr_exact = _make_precision_recall_eval(
+      exact_raw, na_probs, num_true_pos, qid_to_has_ans)
+  pr_f1 = _make_precision_recall_eval(
+      f1_raw, na_probs, num_true_pos, qid_to_has_ans)
+  oracle_scores = {k: float(v) for k, v in qid_to_has_ans.items()}
+  pr_oracle = _make_precision_recall_eval(
+      oracle_scores, na_probs, num_true_pos, qid_to_has_ans)
+  _merge_eval(main_eval, pr_exact, 'pr_exact')
+  _merge_eval(main_eval, pr_f1, 'pr_f1')
+  _merge_eval(main_eval, pr_oracle, 'pr_oracle')
+
+
+def _find_best_thresh(predictions, scores, na_probs, qid_to_has_ans):
+  """Find the best threshold for no answer probability."""
+  num_no_ans = sum(1 for k in qid_to_has_ans if not qid_to_has_ans[k])
+  cur_score = num_no_ans
+  best_score = cur_score
+  best_thresh = 0.0
+  qid_list = sorted(na_probs, key=lambda k: na_probs[k])
+  for qid in qid_list:
+    if qid not in scores: continue
+    if qid_to_has_ans[qid]:
+      diff = scores[qid]
+    else:
+      if predictions[qid]:
+        diff = -1
+      else:
+        diff = 0
+    cur_score += diff
+    if cur_score > best_score:
+      best_score = cur_score
+      best_thresh = na_probs[qid]
+  return 100.0 * best_score / len(scores), best_thresh
+
+
+def _find_all_best_thresh(
+    main_eval, predictions, exact_raw, f1_raw, na_probs, qid_to_has_ans):
+  best_exact, exact_thresh = _find_best_thresh(
+      predictions, exact_raw, na_probs, qid_to_has_ans)
+  best_f1, f1_thresh = _find_best_thresh(
+      predictions, f1_raw, na_probs, qid_to_has_ans)
+  main_eval['final_exact'] = best_exact
+  main_eval['final_exact_thresh'] = exact_thresh
+  main_eval['final_f1'] = best_f1
+  main_eval['final_f1_thresh'] = f1_thresh
+
+
+def evaluate(dataset, predictions, na_probs=None):
+  """Evaluate prediction results."""
+  new_orig_data = []
+  for article in dataset:
+    for p in article['paragraphs']:
+      for qa in p['qas']:
+        if qa['id'] in predictions:
+          new_para = {'qas': [qa]}
+          new_article = {'paragraphs': [new_para]}
+          new_orig_data.append(new_article)
+  dataset = new_orig_data
+
+  if na_probs is None:
+    na_probs = {k: 0.0 for k in predictions}
+  qid_to_has_ans = _make_qid_to_has_ans(dataset)  # maps qid to True/False
+  has_ans_qids = [k for k, v in qid_to_has_ans.items() if v]
+  no_ans_qids = [k for k, v in qid_to_has_ans.items() if not v]
+  exact_raw, f1_raw = _get_raw_scores(dataset, predictions)
+  exact_thresh = _apply_no_ans_threshold(exact_raw, na_probs, qid_to_has_ans)
+  f1_thresh = _apply_no_ans_threshold(f1_raw, na_probs, qid_to_has_ans)
+  out_eval = _make_eval_dict(exact_thresh, f1_thresh)
+  if has_ans_qids:
+    has_ans_eval = _make_eval_dict(
+        exact_thresh, f1_thresh, qid_list=has_ans_qids)
+    _merge_eval(out_eval, has_ans_eval, 'HasAns')
+  if no_ans_qids:
+    no_ans_eval = _make_eval_dict(exact_thresh, f1_thresh, qid_list=no_ans_qids)
+    _merge_eval(out_eval, no_ans_eval, 'NoAns')
+
+  _find_all_best_thresh(
+      out_eval, predictions, exact_raw, f1_raw, na_probs, qid_to_has_ans)
+  _run_precision_recall_analysis(
+      out_eval, exact_raw, f1_raw, na_probs, qid_to_has_ans)
+  return out_eval

official/nlp/data/squad_lib.py

@@ -506,6 +506,34 @@ def write_predictions(all_examples,
   logging.info("Writing predictions to: %s", (output_prediction_file))
   logging.info("Writing nbest to: %s", (output_nbest_file))
 
+  all_predictions, all_nbest_json, scores_diff_json = (
+      postprocess_output(all_examples=all_examples,
+                         all_features=all_features,
+                         all_results=all_results,
+                         n_best_size=n_best_size,
+                         max_answer_length=max_answer_length,
+                         do_lower_case=do_lower_case,
+                         version_2_with_negative=version_2_with_negative,
+                         null_score_diff_threshold=null_score_diff_threshold,
+                         verbose=verbose))
+
+  write_to_json_files(all_predictions, output_prediction_file)
+  write_to_json_files(all_nbest_json, output_nbest_file)
+  if version_2_with_negative:
+    write_to_json_files(scores_diff_json, output_null_log_odds_file)
+
+
+def postprocess_output(all_examples,
+                       all_features,
+                       all_results,
+                       n_best_size,
+                       max_answer_length,
+                       do_lower_case,
+                       version_2_with_negative=False,
+                       null_score_diff_threshold=0.0,
+                       verbose=False):
+  """Postprocess model output, to form predicton results."""
+
   example_index_to_features = collections.defaultdict(list)
   for feature in all_features:
     example_index_to_features[feature.example_index].append(feature)
@@ -676,15 +704,12 @@ def write_predictions(all_examples,
 
     all_nbest_json[example.qas_id] = nbest_json
 
-  with tf.io.gfile.GFile(output_prediction_file, "w") as writer:
-    writer.write(json.dumps(all_predictions, indent=4) + "\n")
+  return all_predictions, all_nbest_json, scores_diff_json
 
-  with tf.io.gfile.GFile(output_nbest_file, "w") as writer:
-    writer.write(json.dumps(all_nbest_json, indent=4) + "\n")
 
-  if version_2_with_negative:
-    with tf.io.gfile.GFile(output_null_log_odds_file, "w") as writer:
-      writer.write(json.dumps(scores_diff_json, indent=4) + "\n")
+def write_to_json_files(json_records, json_file):
+  with tf.io.gfile.GFile(json_file, "w") as writer:
+    writer.write(json.dumps(json_records, indent=4) + "\n")
 
 
 def get_final_text(pred_text, orig_text, do_lower_case, verbose=False):

official/nlp/data/squad_lib_sp.py

@@ -575,10 +575,39 @@ def write_predictions(all_examples,
                       null_score_diff_threshold=0.0,
                       verbose=False):
   """Write final predictions to the json file and log-odds of null if needed."""
-  del do_lower_case, verbose
   logging.info("Writing predictions to: %s", (output_prediction_file))
   logging.info("Writing nbest to: %s", (output_nbest_file))
 
+  all_predictions, all_nbest_json, scores_diff_json = (
+      postprocess_output(all_examples=all_examples,
+                         all_features=all_features,
+                         all_results=all_results,
+                         n_best_size=n_best_size,
+                         max_answer_length=max_answer_length,
+                         do_lower_case=do_lower_case,
+                         version_2_with_negative=version_2_with_negative,
+                         null_score_diff_threshold=null_score_diff_threshold,
+                         verbose=verbose))
+
+  write_to_json_files(all_predictions, output_prediction_file)
+  write_to_json_files(all_nbest_json, output_nbest_file)
+  if version_2_with_negative:
+    write_to_json_files(scores_diff_json, output_null_log_odds_file)
+
+
+def postprocess_output(all_examples,
+                       all_features,
+                       all_results,
+                       n_best_size,
+                       max_answer_length,
+                       do_lower_case,
+                       version_2_with_negative=False,
+                       null_score_diff_threshold=0.0,
+                       verbose=False):
+  """Postprocess model output, to form predicton results."""
+
+  del do_lower_case, verbose
+
   example_index_to_features = collections.defaultdict(list)
   for feature in all_features:
     example_index_to_features[feature.example_index].append(feature)
@@ -740,15 +769,12 @@ def write_predictions(all_examples,
 
     all_nbest_json[example.qas_id] = nbest_json
 
-  with tf.io.gfile.GFile(output_prediction_file, "w") as writer:
-    writer.write(json.dumps(all_predictions, indent=4) + "\n")
+  return all_predictions, all_nbest_json, scores_diff_json
 
-  with tf.io.gfile.GFile(output_nbest_file, "w") as writer:
-    writer.write(json.dumps(all_nbest_json, indent=4) + "\n")
 
-  if version_2_with_negative:
-    with tf.io.gfile.GFile(output_null_log_odds_file, "w") as writer:
-      writer.write(json.dumps(scores_diff_json, indent=4) + "\n")
+def write_to_json_files(json_records, json_file):
+  with tf.io.gfile.GFile(json_file, "w") as writer:
+    writer.write(json.dumps(json_records, indent=4) + "\n")
 
 
 def _get_best_indexes(logits, n_best_size):

official/nlp/optimization.py

@@ -140,19 +140,19 @@ class AdamWeightDecay(tf.keras.optimizers.Adam):
   def apply_gradients(self,
                       grads_and_vars,
                       name=None,
-                      all_reduce_sum_gradients=True):
+                      experimental_aggregate_gradients=True):
     grads, tvars = list(zip(*grads_and_vars))
-    if all_reduce_sum_gradients:
-      # when all_reduce_sum_gradients = False, apply_gradients() no longer
-      # implicitly allreduce gradients, users manually allreduce gradient and
-      # passed the allreduced grads_and_vars. For now, the clip_by_global_norm
-      # will be moved to before the explicit allreduce to keep the math
-      # the same as TF 1 and pre TF 2.2 implementation.
+    if experimental_aggregate_gradients:
+      # when experimental_aggregate_gradients = False, apply_gradients() no
+      # longer implicitly allreduce gradients, users manually allreduce gradient
+      # and passed the allreduced grads_and_vars. For now, the
+      # clip_by_global_norm will be moved to before the explicit allreduce to
+      # keep the math the same as TF 1 and pre TF 2.2 implementation.
       (grads, _) = tf.clip_by_global_norm(grads, clip_norm=1.0)
     return super(AdamWeightDecay, self).apply_gradients(
         zip(grads, tvars),
         name=name,
-        all_reduce_sum_gradients=all_reduce_sum_gradients)
+        experimental_aggregate_gradients=experimental_aggregate_gradients)
 
   def _get_lr(self, var_device, var_dtype, apply_state):
     """Retrieves the learning rate with the given state."""

official/nlp/transformer/misc.py

@@ -239,7 +239,10 @@ def get_callbacks(steps_per_epoch):
   """Returns common callbacks."""
   callbacks = []
   if FLAGS.enable_time_history:
-    time_callback = keras_utils.TimeHistory(FLAGS.batch_size, FLAGS.log_steps)
+    time_callback = keras_utils.TimeHistory(
+        FLAGS.batch_size,
+        FLAGS.log_steps,
+        FLAGS.model_dir if FLAGS.enable_tensorboard else None)
     callbacks.append(time_callback)
 
   if FLAGS.enable_tensorboard:

official/nlp/transformer/transformer_main.py

@@ -246,6 +246,11 @@ class TransformerTask(object):
 
     callbacks = self._create_callbacks(flags_obj.model_dir, 0, params)
 
+    # Only TimeHistory callback is supported for CTL
+    if params["use_ctl"]:
+      callbacks = [cb for cb in callbacks
+                   if isinstance(cb, keras_utils.TimeHistory)]
+
     # TODO(b/139418525): Refactor the custom training loop logic.
     @tf.function
     def train_steps(iterator, steps):
@@ -299,8 +304,13 @@ class TransformerTask(object):
         if not self.use_tpu:
           raise NotImplementedError(
               "Custom training loop on GPUs is not implemented.")
+
         # Runs training steps.
         with summary_writer.as_default():
+          for cb in callbacks:
+            cb.on_epoch_begin(current_iteration)
+            cb.on_batch_begin(0)
+
           train_steps(
               train_ds_iterator,
               tf.convert_to_tensor(train_steps_per_eval, dtype=tf.int32))
@@ -309,10 +319,18 @@ class TransformerTask(object):
           logging.info("Train Step: %d/%d / loss = %s", current_step,
                        flags_obj.train_steps, train_loss)
 
+          for cb in callbacks:
+            cb.on_batch_end(train_steps_per_eval - 1)
+            cb.on_epoch_end(current_iteration)
+
           if params["enable_tensorboard"]:
             for metric_obj in train_metrics:
               tf.compat.v2.summary.scalar(metric_obj.name, metric_obj.result(),
                                           current_step)
+              summary_writer.flush()
+
+        for cb in callbacks:
+          cb.on_train_end()
 
         if flags_obj.enable_checkpointing:
           # avoid check-pointing when running for benchmarking.