Synthia: multidimensional synthetic data generation in Python

Meyer, David E.; Nagler, Thomas

doi:10.21105/joss.02863

Cited by 13 publications

(10 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To improve the results further, the two NNs with the above‐determined configuration are trained with more data. For this, we use Synthia (Meyer & Nagler, 2021) version 0.3.0 (Meyer & Nagler, 2020) as outlined in Meyer, Nagler, et al. (2021) but only for independent inputs.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning Emulation of 3D Cloud Radiative Effects

Meyer

Hogan

Dueben

et al. 2022

J Adv Model Earth Syst

Self Cite

View full text Add to dashboard Cite

The treatment of cloud structure in numerical weather and climate models is often greatly simplified to make them computationally affordable. Here we propose to correct the European Centre for Medium‐Range Weather Forecasts 1D radiation scheme ecRad for 3D cloud effects using computationally cheap neural networks. 3D cloud effects are learned as the difference between ecRad's fast 1D Tripleclouds solver that neglects them and its 3D SPARTACUS (SPeedy Algorithm for Radiative TrAnsfer through CloUd Sides) solver that includes them but is about five times more computationally expensive. With typical errors between 20% and 30% of the 3D signal, neural networks improve Tripleclouds' accuracy for about 1% increase in runtime. Thus, rather than emulating the whole of SPARTACUS, we keep Tripleclouds unchanged for cloud‐free parts of the atmosphere and 3D‐correct it elsewhere. The focus on the comparably small 3D correction instead of the entire signal allows us to improve predictions significantly if we assume a similar signal‐to‐noise ratio for both.

show abstract

Section: Methodsmentioning

confidence: 99%

“…version 0.3.0(Meyer & Nagler, 2020) as outlined in Meyer, Note. Vector quantities are either at the interface between two model layers (half level; HL), or at the model layer (full level; FL).…”

mentioning

confidence: 99%

Machine Learning Emulation of 3D Cloud Radiative Effects

Meyer

Hogan

Dueben

et al. 2022

J Adv Model Earth Syst

Self Cite

View full text Add to dashboard Cite

show abstract

“…Here we use Synthia (Meyer and Nagler, 2020), to fit three different copula types: Gaussian, Vine-parametric, Vinenonparametric. Each copula model is fitted to the training set 𝑋 train in OBT, and to both, 𝑋 train and 𝑌 train sets, in EBT.…”

Section: Methodsmentioning

confidence: 99%

Towards synthetic data generation for machine learning models in weather and climate

Meyer¹

2020

Preprint

Self Cite

View full text Add to dashboard Cite

<p>The use of real data for training machine learning (ML) models are often a cause of major limitations. For example, real data may be (a) representative of a subset of situations and domains, (b) expensive to produce, (c) limited to specific individuals due to licensing restrictions. Although the use of synthetic data are becoming increasingly popular in computer vision, ML models used in weather and climate models still rely on the use of large real data datasets. Here we present some recent work towards the generation of synthetic data for weather and climate applications and outline some of the major challenges and limitations encountered.</p>

show abstract

“…It is noteworthy that the artificial intelligence-based retrosynthesis software Synthia TM failed to identify the U-3CR approach (see Supporting Information, SI). [31,32] Current shortcomings in the field of LAs lie in both their design and their synthesis. [7] First of all, novel compounds and scaffolds are needed with diversity being the focus, as currently a limited number of scaffolds exists.…”

Section: Design Of Amide-caine Type Of Lasmentioning

confidence: 99%

Local Anesthetics via Multicomponent Reactions

et al. 2022

View full text Add to dashboard Cite

Local anesthetics occupy a prime position in clinical medicine as they temporarily relieve the pain by blocking voltage‐gated sodium channels. However, limited structural diversity, problems with the efficiency of syntheses and increasing toxicity, mean that alternative scaffolds with improved chemical syntheses are urgently needed. Here, we demonstrate a multicomponent reaction (MCR)‐based approach both towards the synthesis of commercial local anesthetics and towards novel derivatives as potential anesthesia candidates via scaffold hopping. The reactions are efficient and scalable, and several single‐crystal structures have been obtained. In addition, our methodology has been applied to the synthesis of the antianginal drug ranolazine, via an Ugi three‐component reaction. Representative derivatives from our libraries were evaluated as neuronal activity inhibitors using local field potential recordings (LFPs) in mouse hippocampal brain slices and showed very promising results. This study highlights new opportunities in drug discovery targeting local anesthetics.

show abstract

Synthia: multidimensional synthetic data generation in Python

Cited by 13 publications

References 12 publications

Machine Learning Emulation of 3D Cloud Radiative Effects

Machine Learning Emulation of 3D Cloud Radiative Effects

Towards synthetic data generation for machine learning models in weather and climate

Local Anesthetics via Multicomponent Reactions

Contact Info

Product

Resources

About