add multiple processing

examples/run_squad.py

@@ -360,7 +360,8 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
             doc_stride=args.doc_stride,
             max_query_length=args.max_query_length,
             is_training=not evaluate,
-            return_dataset='pt'
+            return_dataset='pt',
+            threads=args.threads,
         )
 
         if args.local_rank in [-1, 0]:
@@ -478,6 +479,8 @@ def main():
                              "See details at https://nvidia.github.io/apex/amp.html")
     parser.add_argument('--server_ip', type=str, default='', help="Can be used for distant debugging.")
     parser.add_argument('--server_port', type=str, default='', help="Can be used for distant debugging.")
+
+    parser.add_argument('--threads', type=int, default=1, help='multiple threads for converting example to features')
     args = parser.parse_args()
 
     if os.path.exists(args.output_dir) and os.listdir(args.output_dir) and args.do_train and not args.overwrite_output_dir:

transformers/data/processors/squad.py

@@ -4,6 +4,9 @@ import logging
 import os
 import json
 import numpy as np
+from multiprocessing import Pool
+from multiprocessing import cpu_count
+from functools import partial
 
 from ...tokenization_bert import BasicTokenizer, whitespace_tokenize
 from .utils import DataProcessor, InputExample, InputFeatures
@@ -76,47 +79,9 @@ def _is_whitespace(c):
         return True
     return False
 
-def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
-                                       doc_stride, max_query_length, is_training, 
-                                       return_dataset=False):
-    """
-    Converts a list of examples into a list of features that can be directly given as input to a model.
-    It is model-dependant and takes advantage of many of the tokenizer's features to create the model's inputs.
-
-    Args:
-        examples: list of :class:`~transformers.data.processors.squad.SquadExample`
-        tokenizer: an instance of a child of :class:`~transformers.PreTrainedTokenizer`
-        max_seq_length: The maximum sequence length of the inputs.
-        doc_stride: The stride used when the context is too large and is split across several features.
-        max_query_length: The maximum length of the query.
-        is_training: whether to create features for model evaluation or model training.
-        return_dataset: Default False. Either 'pt' or 'tf'.
-            if 'pt': returns a torch.data.TensorDataset,
-            if 'tf': returns a tf.data.Dataset
-
-    Returns:
-        list of :class:`~transformers.data.processors.squad.SquadFeatures`
-
-    Example::
-
-        processor = SquadV2Processor()
-        examples = processor.get_dev_examples(data_dir)
-
-        features = squad_convert_examples_to_features( 
-            examples=examples,
-            tokenizer=tokenizer,
-            max_seq_length=args.max_seq_length,
-            doc_stride=args.doc_stride,
-            max_query_length=args.max_query_length,
-            is_training=not evaluate,
-        )
-    """
-
-    # Defining helper methods    
-    unique_id = 1000000000
-
+def squad_convert_example_to_features(example, max_seq_length,
+                                       doc_stride, max_query_length, is_training):
     features = []
-    for (example_index, example) in enumerate(tqdm(examples)):
     if is_training and not example.is_impossible:
         # Get start and end position
         start_position = example.start_position
@@ -127,8 +92,7 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         cleaned_answer_text = " ".join(whitespace_tokenize(example.answer_text))
         if actual_text.find(cleaned_answer_text) == -1:
             logger.warning("Could not find answer: '%s' vs. '%s'", actual_text, cleaned_answer_text)
-                continue
-
+            return []
 
     tok_to_orig_index = []
     orig_to_tok_index = []
@@ -171,7 +135,8 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
             truncation_strategy='only_second' if tokenizer.padding_side == "right" else 'only_first'
         )
 
-            paragraph_len = min(len(all_doc_tokens) - len(spans) * doc_stride, max_seq_length - len(truncated_query) - sequence_pair_added_tokens)
+        paragraph_len = min(len(all_doc_tokens) - len(spans) * doc_stride,
+                            max_seq_length - len(truncated_query) - sequence_pair_added_tokens)
 
         if tokenizer.pad_token_id in encoded_dict['input_ids']:
             non_padded_ids = encoded_dict['input_ids'][:encoded_dict['input_ids'].index(tokenizer.pad_token_id)]
@@ -202,7 +167,8 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
     for doc_span_index in range(len(spans)):
         for j in range(spans[doc_span_index]["paragraph_len"]):
             is_max_context = _new_check_is_max_context(spans, doc_span_index, doc_span_index * doc_stride + j)
-                index = j if tokenizer.padding_side == "left" else spans[doc_span_index]["truncated_query_with_special_tokens_length"] + j
+            index = j if tokenizer.padding_side == "left" else spans[doc_span_index][
+                                                                   "truncated_query_with_special_tokens_length"] + j
             spans[doc_span_index]["token_is_max_context"][index] = is_max_context
 
     for span in spans:
@@ -224,7 +190,6 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
         # Set the CLS index to '0'
         p_mask[cls_index] = 0
 
-
         span_is_impossible = example.is_impossible
         start_position = 0
         end_position = 0
@@ -251,16 +216,14 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
                 start_position = tok_start_position - doc_start + doc_offset
                 end_position = tok_end_position - doc_start + doc_offset
 
-
         features.append(SquadFeatures(
             span['input_ids'],
             span['attention_mask'],
             span['token_type_ids'],
             cls_index,
             p_mask.tolist(),
-
-                example_index=example_index,
-                unique_id=unique_id,
+            example_index=0,
+            unique_id=0,
             paragraph_len=span['paragraph_len'],
             token_is_max_context=span["token_is_max_context"],
             tokens=span["tokens"],
@@ -269,9 +232,71 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
             start_position=start_position,
             end_position=end_position
         ))
+    return features
 
-            unique_id += 1
+def squad_convert_example_to_features_init(tokenizer_for_convert):
+    global tokenizer
+    tokenizer = tokenizer_for_convert
+
+def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
+                                       doc_stride, max_query_length, is_training, 
+                                       return_dataset=False, threads=1):
+    """
+    Converts a list of examples into a list of features that can be directly given as input to a model.
+    It is model-dependant and takes advantage of many of the tokenizer's features to create the model's inputs.
+
+    Args:
+        examples: list of :class:`~transformers.data.processors.squad.SquadExample`
+        tokenizer: an instance of a child of :class:`~transformers.PreTrainedTokenizer`
+        max_seq_length: The maximum sequence length of the inputs.
+        doc_stride: The stride used when the context is too large and is split across several features.
+        max_query_length: The maximum length of the query.
+        is_training: whether to create features for model evaluation or model training.
+        return_dataset: Default False. Either 'pt' or 'tf'.
+            if 'pt': returns a torch.data.TensorDataset,
+            if 'tf': returns a tf.data.Dataset
+        threads: multiple processing threadsa-smi
+
+
+    Returns:
+        list of :class:`~transformers.data.processors.squad.SquadFeatures`
 
+    Example::
+
+        processor = SquadV2Processor()
+        examples = processor.get_dev_examples(data_dir)
+
+        features = squad_convert_examples_to_features( 
+            examples=examples,
+            tokenizer=tokenizer,
+            max_seq_length=args.max_seq_length,
+            doc_stride=args.doc_stride,
+            max_query_length=args.max_query_length,
+            is_training=not evaluate,
+        )
+    """
+
+    # Defining helper methods    
+    features = []
+    threads = min(threads, cpu_count())
+    with Pool(threads, initializer=squad_convert_example_to_features_init, initargs=(tokenizer,)) as p:
+        annotate_ = partial(squad_convert_example_to_features, max_seq_length=max_seq_length,
+                                       doc_stride=doc_stride, max_query_length=max_query_length, is_training=is_training)
+        features = list(tqdm(p.imap(annotate_, examples, chunksize=32), total=len(examples), desc='convert squad examples to features'))
+    new_features = []
+    unique_id = 1000000000
+    example_index = 0
+    for example_features in tqdm(features, total=len(features), desc='add example index and unique id'):
+        if not example_features:
+            continue
+        for example_feature in example_features:
+            example_feature.example_index = example_index
+            example_feature.unique_id = unique_id
+            new_features.append(example_feature)
+            unique_id += 1
+        example_index += 1
+    features = new_features
+    del new_features
     if return_dataset == 'pt':
         if not is_torch_available():
             raise ImportError("Pytorch must be installed to return a pytorch dataset.")
@@ -296,7 +321,6 @@ def squad_convert_examples_to_features(examples, tokenizer, max_seq_length,
 
         return features, dataset
 
-
     return features