Examples

Various Jupyter notebooks that cover primary usage and peripheral concepts/functionality of xtal2png are included here. Many of these examples can be run on Google Colab and have a corresponding clickable badge at the top of the notebook. For example, here is the badge for the tutorial notebook:

These resources can also be downloaded directly from GitHub (notebooks, scripts).

• Converting interchangeably between crystal structure and a grayscale PNG image (basic xtal2png tutorial)
  • Installation
  • Encode(/decode) two pymatgen Structure objects
  • Visualization
  • Final Remarks
• Selecting Default Parameter Ranges via Materials Project
  • Setup
  • Data
  • Exploratory Data Analysis
  • Selected Parameter Ranges
  • Plotting Histogram Distributions
• Selecting Default Parameter Ranges via Materials Project (Conventional Unit Cell)
  • Setup
  • Data
  • Exploratory Data Analysis
  • Selected Parameter Ranges
  • Plotting Histogram Distributions
• Classification on matbench_mp_is_metal using xtal2png representation of crystal structures
  • Description
  • Benchmark Name
  • Package Versions
  • Algorithm Description
  • Benchmark on Matbench Folds
• XGBoost regression for matbench_mp_e_form task using basic crystallographic features
  • Description
  • Benchmark name
  • Package versions
  • Algorithm description
  • Relevant citations
  • Any other relevant info
  • Running the actual benchmark
  • Check out the results of the benchmark
  • Save our benchmark to file
• Using xtal2png with denoising_diffusion_pytorch
  • MPTimeSplit
  • Convert via xtal2png
  • Denoising Diffusion Probabilistic Model Training
  • Unconditionally Sample New Images
• Using xtal2png with imagen-pytorch
  • Helper
  • Data and xtal2png images
  • Training
  • Sampling
  • deallocate CUDA memory
  • Code Graveyard
• Denoising Diffusion PyTorch Example (script)
• Denoising Diffusion PyTorch Pretrained Sample (script)
• Imagen PyTorch Example (script)