Unified machine learning protocol for copolymer structure-property predictions.

Unified machine learning protocol for copolymer structure-property predictions.
复制标题

DOI:
10.1016/j.xpro.2022.101875
复制
发表时间:
2022-12-16
期刊:
影响因子:
--
通讯作者:
Li, Ying
Li, Ying
中科院分区:
其他
文献类型:
--
作者:
Tao, Lei;Arbaugh, Tom;Byrnes, John;Varshney, Vikas;Li, Ying

文献摘要

参考文献

相似文献

在预测聚合物性能时,结构 - 性能关系极具价值。本实验方案展示了一种基于多种机器学习(ML)架构的分步式方法,该方法能够处理诸如交替共聚物、无规共聚物、嵌段共聚物和梯度共聚物等共聚物类型。我们详细说明了必要软件的安装步骤以及数据集的构建方法。我们进一步阐述了四种神经网络模型的训练与优化步骤,以及随后利用训练值和测试值对模型进行可视化和比较的过程。 如需了解本实验方案使用及实施的完整细节,请参考陶等人(2022年)的研究。 构建共聚物机器学习模型的详细步骤 兼顾共聚物的化学组成和序列分布 运用四种机器学习模型分析不同类型的共聚物 区分无规、嵌段和梯度共聚物的序列模式 出版方声明:进行任何实验方案都需遵循当地机构关于实验室安全和伦理的指导方针。 在预测聚合物性能时,结构 - 性能关系极具价值。本实验方案展示了一种基于多种机器学习(ML)架构的分步式方法,该方法能够处理诸如交替共聚物、无规共聚物、嵌段共聚物和梯度共聚物等共聚物类型。我们详细说明了必要软件的安装步骤以及数据集的构建方法。我们进一步阐述了四种神经网络模型的训练与优化步骤,以及随后利用训练值和测试值对模型进行可视化和比较的过程。
Structure-property relationships are extremely valuable when predicting the properties of polymers. This protocol demonstrates a step-by-step approach, based on multiple machine learning (ML) architectures, which is capable of processing copolymer types such as alternating, random, block, and gradient copolymers. We detail steps for necessary software installation and construction of datasets. We further describe training and optimization steps for four neural network models and subsequent model visualization and comparison using training and test values. For complete details on the use and execution of this protocol, please refer to Tao et al. (2022). Detailed steps for building machine learning model for copolymers Consideration of both chemical composition and sequence distribution of copolymers Analysis of different copolymers types using four machine learning models Differentiation of sequence patterns of random, block, and gradient copolymers Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics. Structure-property relationships are extremely valuable when predicting the properties of polymers. This protocol demonstrates a step-by-step approach, based on multiple machine learning (ML) architectures, which is capable of processing copolymer types such as alternating, random, block, and gradient copolymers. We detail steps for necessary software installation and construction of datasets. We further describe training and optimization steps for four neural network models and subsequent model visualization and comparison using training and test values.
DOI: 10.1016/j.patter.2021.100225
发表时间: 2021-04-09
期刊: Patterns (New York, N.Y.)
影响因子: --
作者:
Tao L;Chen G;Li Y
通讯作者: Li Y
DOI: 10.1021/c160017a018
发表时间: 1965-01-01
期刊: JOURNAL OF CHEMICAL DOCUMENTATION
影响因子: --
作者:
MORGAN, HL
通讯作者: MORGAN, HL
DOI: 10.1021/jacs.1c08181
发表时间: 2021-10-12
影响因子: 15
作者:
Reis, Marcus;Gusev, Filipp;Leibfarth, Frank A.
通讯作者: Leibfarth, Frank A.
DOI: 10.1039/c8sc05710a
发表时间: 2019-05-21
期刊: CHEMICAL SCIENCE
影响因子: 8.4
作者:
Wilbraham, Liam;Sprick, Reiner Sebastian;Zwijnenburg, Martijn A.
通讯作者: Zwijnenburg, Martijn A.
DOI: 10.1021/acs.jcim.1c01031
发表时间: 2021-11-22
影响因子: 5.6
作者:
Tao, Lei;Varshney, Vikas;Li, Ying
通讯作者: Li, Ying