Skip to content

Latest commit

 

History

History
35 lines (28 loc) · 1.56 KB

README.md

File metadata and controls

35 lines (28 loc) · 1.56 KB

Dataset Preparation

Data Source

We use The Material Project as the source of the dataset generation.

Usage

Prepare API_KEY for The Material Project

  1. Create an account: https://profile.materialsproject.org/
  2. Visit and generate your API_KEY: https://next-gen.materialsproject.org/api#api-key
  3. Save your API_KEY as "API_KEY.txt" in this directory

Running Scripts

  • The following commands assume you are running in /workspace in the Docker environment.
  • The working directory is /workspace/data and finally /workspace/data/dataset/* will be created.
  1. Download The Material Project data.
    • poetry run python src/dataset_generation/01_download.py.
  2. Extract structures from the crawled data.
    • poetry run python src/dataset_generation/02_extract_structures.py.
  3. Convert to a conventional cell. -poetry run python src/dataset_generation/03_convert_to_conventional_cell.py.
  4. Select materials according to each criterion.
    • poetry run python src/dataset_generation/04_01_select_materials_by_cell.py.
      • For the lim_l6 dataset.
    • poetry run python src/dataset_generation/04_02_select_materials_by_formula.py.
      • For the ICSG3D dataset.
    • Run python src/dataset_generation/04_03_select_materials_by_list.py.
      • For the YBCO13 dataset.
  5. Split the dataset into train/validation/test splits.
    • poetry run python src/dataset_generation/05_split.py.
  6. Generate the NeSF style dataset.
    • poetry run python src/dataset_generation/06_generate_NeSF_dataset.py.