onehaitao/att-blstm-relation-extraction

Implementation of Attention-Based Bidirectional Long Short-Term Memory Networks for Relation Classification.

Hi there 👋

My name is Haitao Wang.

• 👻 An ordinary coder, enjoy life and love technology.
• ⚽ Interested in information extraction and natural language understanding.
• ✉️ How to reach me: Issue or Mail.

att-blstm-relation-extraction's Issues

how to deal with the problem aboutAttributeError: 'SemEvalDataLoader' object has no attribute '__collate_fn'

AttributeError: 'SemEvalDataLoader' object has no attribute '__collate_fn'
I want to know how to deal with it ,where is the '__collate_fn'

I run the code and get the F1 score 82.73

Hi, thanks for your code!I have a question,can you give me some advice? I run your code,but only get F1 score 82.73,never get 83.13 in your log file.If you have time, thanks for your advice.

Position Indicator

Hello,
I wonder if the position indicator has the corresponding embedding or just UNK?

Thanks

当选择BiLSTtt进行训练时，报错

往上翻发现提示
File "D:\Program Files\miniconda3\envs\pyt\lib\site-packages\torch\nn\modules\module.py", line 1072, in train
module.train(mode)
[Previous line repeated 990 more times]
File "D:\Program Files\miniconda3\envs\pyt\lib\site-packages\torch\nn\modules\module.py", line 1070, in train
self.training = mode
File "D:\Program Files\miniconda3\envs\pyt\lib\site-packages\torch\nn\modules\rnn.py", line 97, in setattr
super(RNNBase, self).setattr(attr, value)
File "D:\Program Files\miniconda3\envs\pyt\lib\site-packages\torch\nn\modules\module.py", line 602, in setattr
params = self.dict.get('_parameters')
RecursionError: maximum recursion depth exceeded while calling a Python object

Precison and Recall Misplaced

If rows of confusion matrix are prediction and columns are labels, it seems they are misplaced.

Att-BLSTM-relation-extraction/evaluate.py

Line 31 in 90db246

p = float(confusion_matrix[i][i]) / float(col_sum[i] + xDIRx[i])

运行发现错误

---

start to train the model ...
Traceback (most recent call last):
File "", line 1, in
File "C:\Program Files\Python36\lib\multiprocessing\spawn.py", line 105, in spawn_main
exitcode = _main(fd)
File "C:\Program Files\Python36\lib\multiprocessing\spawn.py", line 115, in _main
self = reduction.pickle.load(from_parent)
AttributeError: 'SemEvalDataLoader' object has no attribute '__collate_fn'
Traceback (most recent call last):
File "", line 1, in
File "C:\Program Files\Python36\lib\multiprocessing\spawn.py", line 105, in spawn_main
exitcode = _main(fd)
File "C:\Program Files\Python36\lib\multiprocessing\spawn.py", line 115, in _main
self = reduction.pickle.load(from_parent)
AttributeError: 'SemEvalDataLoader' object has no attribute '__collate_fn'
Traceback (most recent call last):
File "run.py", line 100, in
train(model, criterion, loader, config)
File "run.py", line 37, in train
for step, (data, label) in enumerate(train_loader):
File "C:\Program Files\Python36\lib\site-packages\torch\utils\data\dataloader.py", line 804, in next
idx, data = self._get_data()
File "C:\Program Files\Python36\l

semeval_scorer()计算分数问题

您好，我看代码中的该函数计算分数似乎不关注标签的(e1,e2)的位置，这一点有一些疑惑。

TypeError: __collate_fn()

python = 3.6
pytorch = 1.3.0
After python run.py, show this error massage

Traceback (most recent call last):
File ".\run.py", line 105, in
train(model, criterion, loader, config)
File ".\run.py", line 39, in train
for step, (data, label) in enumerate(train_loader):
File "C:\Users\90607.conda\envs\Att-BLSTM-relation-extraction-master\lib\site-packages\torch\utils\data\dataloader.py", line 346, in next
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
File "C:\Users\90607.conda\envs\Att-BLSTM-relation-extraction-master\lib\site-packages\torch\utils\data_utils\fetch.py", line 47, in fetch
return self.collate_fn(data)
TypeError: __collate_fn() takes 1 positional argument but 2 were given