SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

add is_impossible tensor to model inputs during fine-tuning xlnet on squad2.0

Browse Source
This commit is contained in:
Xu Hongshen
2019-11-12 16:11:24 +08:00
parent 74ce8de7d8
commit 7da3ef24cd
1 changed files with 4 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
examples/run_squad.py
View File

@@ -147,6 +147,8 @@ def train(args, train_dataset, model, tokenizer):
if args.model_type in ['xlnet', 'xlm']: if args.model_type in ['xlnet', 'xlm']:
inputs.update({'cls_index': batch[5], inputs.update({'cls_index': batch[5],
'p_mask': batch[6]}) 'p_mask': batch[6]})
if args.version_2_with_negative:
inputs.update({'is_impossible': batch[7]})
outputs = model(**inputs) outputs = model(**inputs)
loss = outputs[0] # model outputs are always tuple in transformers (see doc) loss = outputs[0] # model outputs are always tuple in transformers (see doc)
@@ -339,9 +341,10 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
else: else:
all_start_positions = torch.tensor([f.start_position for f in features], dtype=torch.long) all_start_positions = torch.tensor([f.start_position for f in features], dtype=torch.long)
all_end_positions = torch.tensor([f.end_position for f in features], dtype=torch.long) all_end_positions = torch.tensor([f.end_position for f in features], dtype=torch.long)
all_is_impossible = torch.tensor([1.0 if f.is_impossible == True else 0.0 for f in features], dtype=torch.float)
dataset = TensorDataset(all_input_ids, all_input_mask, all_segment_ids, dataset = TensorDataset(all_input_ids, all_input_mask, all_segment_ids,
all_start_positions, all_end_positions, all_start_positions, all_end_positions,
all_cls_index, all_p_mask) all_cls_index, all_p_mask, all_is_impossible)
if output_examples: if output_examples:
return dataset, examples, features return dataset, examples, features