[RE_BGRU_2ATT]Data directory missing?

[juncaofish]

Just confirm if this folder is not provided.

[crownpku]

The original relationid, training and test data samples and word embedding are in RE_BGRU_2ATT/origin_data. You may provide your own training data according to the same format.

Then as mentioned in readme you have to run

#python3 initial.py

to generate .npy files that will be saved at /data folder.

[juncaofish]

Thanks! Will give it a try.

[crownpku]

If you need more data for training this task you may check out a larger corpus here.