From 794bb3610a28296bc78a56fd30f0f3b0801e4f91 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jos=C3=A9=20Manuel=20N=C3=A1poles=20Duarte?= Date: Fri, 29 Mar 2024 00:58:11 -0600 Subject: [PATCH] Update README.md --- README.md | 43 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 43 insertions(+) diff --git a/README.md b/README.md index 247d278..4622c9a 100644 --- a/README.md +++ b/README.md @@ -24,6 +24,49 @@ data resources. This endeavor is crucial for maximizing the efficacy of computat constrained, ensuring that Bayesian optimization remains a viable and effective tool for advancing research and applications in deep learning for chemistry and beyond. +Results: +| N | Learning Rate | Loss | +|------|------------------|-----------------| +| 200 | 0.084878 | 0.010064 | +| 400 | 0.059692 | 0.003261 | +| 600 | 0.055656 | 0.003610 | +| 800 | 0.029240 | 0.001820 | +| 1000 | 0.017758 | 0.001970 | +| 1200 | 0.027993 | 0.003351 | +| 1400 | 0.018717 | 0.002034 | +| 1600 | 0.023550 | 0.002136 | +| 1800 | 0.014614 | 0.002025 | +| 2000 | 0.009610 | 0.001134 | + +| N | Batch Size | Loss | +|------|--------|------------| +| 200 | 120 | 0.0080974 | +| 400 | 21 | 0.002841 | +| 600 | 27 | 0.002955 | +| 800 | 16 | 0.002613 | +| 1000 | 22 | 0.002506 | +| 1200 | 22 | 0.002462 | +| 1400 | 30 | 0.001593 | +| 1600 | 16 | 0.002007 | +| 1800 | 18 | 0.002203 | +| 2000 | 16 | 0.002041 | + +| N | # of Neurons | Loss | +|------|--------|------------| +| 200 | 111 | 0.0084131 | +| 400 | 47 | 0.0043065 | +| 600 | 51 | 0.002910 | +| 800 | 32 | 0.002574 | +| 1000 | 22 | 0.002193 | +| 1200 | 34 | 0.002597 | +| 1400 | 22 | 0.002111 | +| 1600 | - | - | +| 1800 | 48 | 0.001758 | +| 2000 | 59 | 0.001668 | + + + + References: 1. Anatole von Lilienfeld and Kieron Burke. “Retrospective on a decade of machine learning for chemical discovery”. In: Nature Communications 11.1 (Sept. 2020). DOI: 10.1038/s41467- 020- 18556- 9. URL: https: