clairett/deep-memory-network

Implement deep memory network used for Aspect Level Sentiment Classification

Aspect Level Sentiment Classification with Deep Memory Network

TensorFlow implementation of Tang et al.'s EMNLP 2016 work.

Problem Statement

Given a sentence and an aspect occurring in the sentence, this task aims at inferring the sentiment polarity (e.g. positive, negative, neutral) of the aspect.

Example

For example, in sentence ''great food but the service was dreadful!'', the sentiment polarity of aspect ''food'' is positive while the polarity of aspect ''service'' is negative.

Quick Start

Install this quick GLOVE embeddings loading tool

Runs on python3 and tensorflow 1.4.1

Train a model with 3 hops on the Restaurant dataset.

python main.py --show True

Performance

Achieved accuracy of 72% for Laptop and 79% for Restaurant.

Acknowledgements

• More than 80% of the code is borrowed from ganeshjawahar.
• Using this code means you have read and accepted the copyrights set by the dataset providers.

Author

Tian Tian

Licence

MIT

deep-memory-network's Issues

AttributeError: 'array.array' object has no attribute 'tobytes'

Thanks for sharing the code,
I am using python 2, got this error:
using coding utf-8, otherwise will got another coding error.

import sys

reload(sys)
sys.setdefaultencoding('utf8')

(deepMemNet) zchen@hdp-node11:~/Deep-Memory-Network$ python main.py --show True
/home/zchen/anaconda2/lib/python2.7/site-packages/h5py/init.py:36: FutureWarning: Conversion of the second argument of issubdtype from float to np.floating is deprecated. In future, it will be treated as np.float64 == np.dtype(float).type.
from ._conv import register_converters as _register_converters
0%| | 778/2195895 [00:00<09:25, 3885.03it/s]Traceback (most recent call last):
File "main.py", line 73, in
tf.app.run()
File "/home/zchen/anaconda2/lib/python2.7/site-packages/tensorflow/python/platform/app.py", line 124, in run
_sys.exit(main(argv))
File "main.py", line 46, in main
embeddings = init_word_embeddings(FLAGS.pretrain_embeddings, word_set, FLAGS.edim)
File "/home/zchen/Deep-Memory-Network/data.py", line 16, in init_word_embeddings
embedding = GloveEmbedding(tokens[1], d_emb=edim, show_progress=True)
File "/home/zchen/anaconda2/lib/python2.7/site-packages/embeddings/glove.py", line 48, in init
self.load_word2emb(show_progress=show_progress)
File "/home/zchen/anaconda2/lib/python2.7/site-packages/embeddings/glove.py", line 81, in load_word2emb
self.insert_batch(batch)
File "/home/zchen/anaconda2/lib/python2.7/site-packages/embeddings/embedding.py", line 139, in insert_batch
binarized = [(word, array('f', emb).tobytes()) for word, emb in batch]
AttributeError: 'array.array' object has no attribute 'tobytes'
Exception KeyError: KeyError(<weakref at 0x7f70f9d98b50; to 'tqdm' at 0x7f70f9cc1c10>,) in <bound method tqdm.del of 0%| | 778/2195895 [00:00<24:56, 1467.16it/s]> ignored