Hello! ✌️ I'm Marcel, a postdoc at EPFL in Lausanne, working somewhere in between machine learning and computational condensed matter physics. You can find more about my work here, or you could say say hi on twitter.
This is where I keep my notes on work-related things. Maybe they'll be useful for you!
(This website is rather new, and probably a bit wonky in places – if anything suspicious happens, please drop me a line at mail@marcel.science, or hit me up on twitter.)
Notes
- numpy.load access performance
- torch.nn.Linear initialisation
- torch gradcheck and pytest
- ase.lol
- JAX-MD Neighbourlists in Non-Orthorhombic Systems
- Removing ticks in log scale
- Stress and PaiNN
- Indexing arrays with other arrays
- Transposing a 'sparse', randomly-ordered array
- Interfacing ase and i-pi
- np.loadtxt from string