(Unofficial) Pytorch Implementation of "Adaptive Co-attention Network for Named Entity Recognition in Tweets" (AAAI 2018)

• python>=3.5
• torch==1.3.1
• torchvision==0.4.2
• pillow==7.0.0
• pytorch-crf==0.7.2
• seqeval==0.0.12
• gdown>=3.10.1

$ pip3 install -r requirements.txt

• Original code's pretrained word embedding can be downloaded at here.
• But it takes quite a long time to download, so I take out the word vectors (word_vector_200d.vec) that are only in word vocab.

• Image features are extracted from last pooling layer of VGG16.

• If you want to extract the feature by yourself, follow as below.

  1. Clone the repo of original code.
  2. Copy data/ner_img from original code to this repo.
  3. Run as below. (img_vgg_features.pt will be saved in data dir)

  $ python3 save_vgg_feature.py

• Extracted features will be downloaded automatically when you run main.py.

• There are some differences between the paper and the original code, so I tried to follow the paper's equations as possible.
• Build the vocab with train, dev, and test dataset. (same as the original code)
  • Making the vocab only with train dataset decreases performance a lot. (about 5%)
• Use Adam optimizer instead of RMSProp.

$ python3 main.py --do_train --do_eval

• Original Paper
• Original Code Implementation