Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Unable to reproduce results from the paper #15

Open
dheerajrajagopal opened this issue Dec 26, 2017 · 3 comments
Open

Unable to reproduce results from the paper #15

dheerajrajagopal opened this issue Dec 26, 2017 · 3 comments

Comments

@dheerajrajagopal
Copy link

Hi,
I wanted to reproduce results from table 1 from the paper (BioASQ only, Experiment (1) ). I have been using the default settings from the README file. This is my setting

python biomedical_qa/training/train_qa.py
--with_question_type_features
--size 100
--max_epochs 40
--model_type simple_pointer
--is_bioasq True
--data data/bioasq_v2_cv_4 \ # 4th cross-validation set
--trainset_prefix train
--validset_prefix dev
--start_output_unit sigmoid
--ckpt_its 10
--batch_size 64
--save_dir saved_models_v3_cv_4 \ # directory for cross validation set 4
--dropout 0.5
--dataset squad
--transfer_model_config model_checkpoints/glove_pubmed_embedder/config.pickle
--transfer_model_path model_checkpoints/glove_pubmed_embedder/model.tf
--composition LSTM

I am unable to reproduce the numbers from the paper. I was wondering if I am doing something wrong in the settings.

Could you help me figure out ?

@Chunlinx
Copy link

Hi,
When I try to train your project , I got an error:
ModuleNotFoundError: No module named 'web.embeddings'

could you tell me why this happend?

@Ali-Xain
Copy link

@Chunlinx if you are using Python 3+ just replace import web.embeddings with import web
but first install web module in your python environment.

@lbda1
Copy link

lbda1 commented Jan 20, 2020

pip install web
but there is no such this package

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants