/WSM_project2

網路搜索與探勘 專案2

Primary LanguagePython

WSM project 2

Quick Start

  1. Install Poetry 
    curl -sSL https://install.python-poetry.org | python3 -
  2. Use Python 3.12 to run the project 
    pyenv local 3.12  # if you have pyenv installed
poetry env use $(command -v python3)  # make sure the right python is used
  3. Install dependencies 
    poetry install --no-root
  4. Put in WT2G corpus in root and proj2_sample_run/data folder
  5. Index the corpus 
    ./indexing.sh
  6. Run the programs. The relation of them is shown below:
    • Queries part 1: Run search.py with different parameters. Currently ran results are in runs folder.
    • Queries part 2: Run queries_part2/train.py to train the model. The trained model is in model.pth. The performance of the trained model will be calculated. Meanwhile, run queries_part2/eval_part1.py to evaluate the performance of those ranking functions in part 1.
    • Evaluation part 1: Run search.py with different parameters. Currently ran results are in eval_part1 folder.
    • Evaluation part 2: Run search.py with different parameters to see the results from those three ranking functions. And then, run eval_part2/eval.py to calculate the performance of the model. Currently ran results are in eval_part2 folder.