Skip to content

Latest commit

 

History

History
23 lines (20 loc) · 1.4 KB

CONTRIBUTING.md

File metadata and controls

23 lines (20 loc) · 1.4 KB
Raw

Contributing to tool recommender system in Galaxy using deep learning (Gated recurrent units neural network)

Following steps can be followed to start contributing to this project:

  1. Fork this repository (https://github.com/anuprulez/galaxy_tool_recommendation).
  2. Create a new branch.
  3. Install the dependencies by executing the following lines:
    • conda env create -f environment.yml
    • conda activate tool_prediction_gru_wc
  4. The scripts are located at scripts/.
  5. Data files is located at data/.
  6. Add new features/techniques.
  7. Run the project using sh train.sh.
    • To run this project on complete set of workflow, large compute resource is needed (with at least 20-30 GB RAM) and running time is > 24 hours.
    • Details of the parameters in the training script are given in README.md.
  8. Get a recommended model at data/<<file name>>.hdf5.
  9. See recommended tools using ipython_script/tool_recommendation_gru_wc.ipynb or place the newly created recommendation model (from step 7) at ipython_script/data/<<file name>>.hdf5.
  10. Open a pull request against the main repository (https://github.com/anuprulez/galaxy_tool_recommendation).

Contributors

  1. Anup Kumar (https://github.com/anuprulez) (Main contributor).
  2. Helena Rasche (https://github.com/hexylena) (Contributed to the scripts for data collection from Galaxy EU server and to the Galaxy tool prediction API).
  3. ...