SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2311.01135 · NeurIPS · 2023

Generating QM1B with PySCF IPU

Dominique Beaini, Carlo Luschi, Dominic Masters, Hatem Helal, Andrew Fitzgibbon, Alexander Mathiasen, Kerstin Klaser, Paul Balanca, Josef Dean, Dft Datasets

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 3 functions out of this paper's own repositories and ran 2 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
graphcore-research/pyscf-ipu canonical 1 of 2
graphcore-research/qm1b-dataset canonical 1 of 1
FunctionStatusWhere it lives
f Ran graphcore-research/pyscf-ipu/pyscf_ipu/dft.py
code served (permissive licence) · get_code("554977c240abe619")
md5 Ran graphcore-research/qm1b-dataset/download.py
code served (permissive licence) · get_code("6e02f22bcaa527c3")
get_atom_string Not yet run graphcore-research/pyscf-ipu/pyscf_ipu/dft.py
code served (permissive licence) · get_code("1e11ba21ba669e92")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

The emergence of foundation models in Computer Vision and Natural Language Processing have resulted in immense progress on downstream tasks. This progress was enabled by datasets with billions of training examples. Similar benefits are yet to be unlocked for quantum chemistry, where the potential of deep learning is constrained by comparatively small datasets with 100k to 20M training examples. These datasets are limited in size because the labels are computed using the accurate (but computationally demanding) predictions of Density Functional Theory (DFT). Notably, prior DFT datasets were created using CPU supercomputers without leveraging hardware acceleration. In this paper, we take a first step towards utilising hardware accelerators by introducing the data generator PySCF IPU using Intelligence Processing Units (IPUs). This allowed us to create the dataset QM1B with one billion training examples containing 9-11 heavy atoms. We demonstrate that a simple baseline neural network (SchNet 9M) improves its performance by simply increasing the amount of training data without additional inductive biases. To encourage future researchers to use QM1B responsibly, we highlight several limitations of QM1B and emphasise the low-resolution of our DFT options, which also serves as motivation for even larger, more accurate datasets. Code and dataset.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2311.01135")
get_code_for_paper("2311.01135")
have("2311.01135")

Connect an agent — have() is free.