deep_srl
deep_srl copied to clipboard
luheng
Code and pre-trained model for: Deep Semantic Role Labeling: What Works and What's Next
Hi, Luheng: Thanks for your great work! I encountered some strange errors during training. I used the following to start training your model : `python python/train.py --config=./config/srl_config.json --model=./output --train=./sample_data/sentences_with_gold.txt --dev=./sample_data/sentences_with_gold.txt...
@luheng Did you use the complete ontonotes 5.0 dataset or CoNLL 2012 train/test/dev split ?
Needed for PERL5LIB and correct PATH variables, used later in the script.
Dear author: I have some questions for the data, the conll2012 data downloaded from the [website: V12 release](http://cemantix.org/data/ontonotes.html) you supplied could not being processed by its own [process scripts](http://ontonotes.cemantix.org/download/conll-formatted-ontonotes-5.0-scripts.tar.gz) due...
(.venv) ub16hp@UB16HP:~/ub16_prj/deep_srl$ bash scripts/run_train.sh 0 Traceback (most recent call last): File "python/train.py", line 4, in from neural_srl.theano.tagger import BiLSTMTaggerModel File "/home/ub16hp/ub16_prj/deep_srl/python/neural_srl/theano/tagger.py", line 1, in from optimizer import * File "/home/ub16hp/ub16_prj/deep_srl/python/neural_srl/theano/optimizer.py",...
Hello, First thank you very much for this contribution! However, when I try to train the model from scratch I'm getting, lower performance. (about 1pt of f-score) I'm able to...
1.why the file sentences_with_predicates.txt and sentences_without_predicates.txt is the same? 2.what the 'gold' word mean(like,data/srl/conll05.devel.props.gold.txt) ? Is't means predicates?
Error message: in File "/root/test-deepsrl/python/neural_srl/shared/conll_utils.py", line 29, in print_sentence_to_conll assert len(label_column) == len(tokens) model: conll_2012_model input file: see attachment. [temp.txt](https://github.com/luheng/deep_srl/files/1726169/temp.txt)
