Delete Python3.5 incompatible multiprocessing.Pool usage of BERTDatasetTransform

Traceback (most recent call last):
  File "./scripts/bert/finetune_classifier.py", line 373, in <module>
    bert_tokenizer, task, batch_size, dev_batch_size, args.max_len, vocabulary, args.pad)
  File "./scripts/bert/finetune_classifier.py", line 306, in preprocess_data
    data_train = mx.gluon.data.SimpleDataset(pool.map(trans, train_tsv))
  File "/var/lib/jenkins/workspace/gluon-nlp-gpu-py3-master@6/conda/gpu/py3-master/lib/python3.5/multiprocessing/pool.py", line 266, in map
    return self._map_async(func, iterable, mapstar, chunksize).get()
  File "/var/lib/jenkins/workspace/gluon-nlp-gpu-py3-master@6/conda/gpu/py3-master/lib/python3.5/multiprocessing/pool.py", line 644, in get
    raise self._value
  File "/var/lib/jenkins/workspace/gluon-nlp-gpu-py3-master@6/conda/gpu/py3-master/lib/python3.5/multiprocessing/pool.py", line 424, in _handle_tasks
    put(task)
  File "/var/lib/jenkins/workspace/gluon-nlp-gpu-py3-master@6/conda/gpu/py3-master/lib/python3.5/multiprocessing/connection.py", line 206, in send
    self._send_bytes(ForkingPickler.dumps(obj))
  File "/var/lib/jenkins/workspace/gluon-nlp-gpu-py3-master@6/conda/gpu/py3-master/lib/python3.5/multiprocessing/reduction.py", line 50, in dumps
    cls(buf, protocol).dump(obj)
_pickle.PicklingError: Can't pickle <class 'module'>: attribute lookup module on builtins failed

Author: leezu

scripts/bert/finetune_classifier.py

@@ -39,7 +39,6 @@
 import random
 import logging
 import warnings
-import multiprocessing
 import numpy as np
 import mxnet as mx
 from mxnet import gluon
@@ -289,8 +288,6 @@
 
 def preprocess_data(tokenizer, task, batch_size, dev_batch_size, max_len, vocab, pad=False):
     """Train/eval Data preparation function."""
-    pool = multiprocessing.Pool()
-
     # transformation for data train and dev
     label_dtype = 'float32' if not task.class_labels else 'int32'
     trans = BERTDatasetTransform(tokenizer, max_len,
@@ -303,7 +300,7 @@ def preprocess_data(tokenizer, task, batch_size, dev_batch_size, max_len, vocab,
     # data train
     # task.dataset_train returns (segment_name, dataset)
     train_tsv = task.dataset_train()[1]
-    data_train = mx.gluon.data.SimpleDataset(pool.map(trans, train_tsv))
+    data_train = mx.gluon.data.SimpleDataset(list(map(trans, train_tsv)))
     data_train_len = data_train.transform(
         lambda input_id, length, segment_id, label_id: length, lazy=False)
     # bucket sampler for training
@@ -331,7 +328,7 @@ def preprocess_data(tokenizer, task, batch_size, dev_batch_size, max_len, vocab,
     dev_tsv_list = dev_tsv if isinstance(dev_tsv, list) else [dev_tsv]
     loader_dev_list = []
     for segment, data in dev_tsv_list:
-        data_dev = mx.gluon.data.SimpleDataset(pool.map(trans, data))
+        data_dev = mx.gluon.data.SimpleDataset(list(map(trans, data)))
         loader_dev = mx.gluon.data.DataLoader(
             data_dev,
             batch_size=dev_batch_size,
@@ -356,7 +353,7 @@ def preprocess_data(tokenizer, task, batch_size, dev_batch_size, max_len, vocab,
     test_tsv_list = test_tsv if isinstance(test_tsv, list) else [test_tsv]
     loader_test_list = []
     for segment, data in test_tsv_list:
-        data_test = mx.gluon.data.SimpleDataset(pool.map(test_trans, data))
+        data_test = mx.gluon.data.SimpleDataset(list(map(test_trans, data)))
         loader_test = mx.gluon.data.DataLoader(
             data_test,
             batch_size=dev_batch_size,