Skip to content

Winning solution for hackerearth's deep learning #4 challenge

Notifications You must be signed in to change notification settings

kunwar31/hackerearth-tag-recommendation

Repository files navigation

Requirements

  • Jupyter notebook
  • Keras
  • Tensorflow
  • CUDA capable GPU ( Replace CuDNNGRU/CuDNNLSTM with GRU/LSTM if not available )
  • Pandas
  • SciKit Learn
  • Numpy
  • TQDM

Training Workflow

Please run these jupyter notebooks in order in order to generate the submission file.

  1. parse_html
  2. clean_data
  3. clean_data_punc_brute
  4. embeddings
  5. embeddings_punc_brute
  6. 1_512
  7. 512_1024
  8. 1024_1536
  9. 1536_2048
  10. 2048_2560
  11. 2560_3072
  12. Combine

submission file : final_submission.csv

Solution overview

Externel Resources/ Citation

All (Most) of the externel resources such as code snippets, embeddings, etc. have been linked to their sources in the notebooks itself.

About

Winning solution for hackerearth's deep learning #4 challenge

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published