1) training set
训练集
1.
Effects of category distribution in a training set on text categorization;
训练集类别分布对文本分类的影响
2.
A simplification SVM model based on guard vectors is proposed for overcoming the slow speed of training and classification for large scale training set.
针对支持向量机(SVM)在处理大规模训练集时,训练速度和分类速度变慢的缺点,提出了一种基于卫向量的简化SVM模型。
3.
For this method,in the training process,a simple classifier based on the idea of each class having a center is designed,which can select efficient data from plenty of unlabeled data to label and then add into the SVM training set.
该方法在训练过程中,先构造一个基于类中心思想的简易分类器,通过设定有效阈值,从未标记数据中挑选区别度较大的数据加入到SVM的训练集中;在分类过程中,根据待分类点与分类面的相对位置,结合SVM和KNN算法,分两种情况来对其进行分类。
2) concentration training
集中训练
1.
The key for implementing professional skills identification lies in taking modularized curriculum system as the framework,integrating skills identification with diploma education examination,adopting the method of combination of curriculum alternation,concentration training with examination for certificate,and strengthening the cultivation of students operational abilities.
实施职业技能鉴定的关键在于以模块式课程体系为框架,将技能鉴定与学历教育考核融为一体,采用课程穿插、集中训练、考证相结合的方法,强化学生动手能力培养。
3) integration trainingt
集成训练
4) training subset
训练子集
1.
Error correlation,which acts as a parameter to select the training subset from training set,is developed to improve the generalizability of learning machine.
针对有限样本学习机器的偏差/方差的困境,以及过拟合引起的泛化性能的下降,分析了样本选择对学习机器泛化的影响,提出误差相关度学习算法ECL,利用误差相关度来权衡偏差和方差的关系,避免了求解复杂学习系统的VC维数,并以样本点的误差相关度为指标来选择训练子集,提高学习机器的泛化性能。
5) group training
集体训练
1.
The study of the influence of group training on moral traits of undergraduates shows that the cohesiveness index is significantly improved after training.
集体训练对大学生品德影响的相关研究结果显示:群体凝聚力指数在集体训练后有显著提高;在协变考虑对照组的前提下,“人性哲学量表”上的“利他”、“意志力与理性”、“独立性”几项分量表也都有显著变化。
2.
Those in test group received group training with self-designed community based motor function rehabilitation simple technique combi.
方法:采用小样本、随机对照研究方法,选取北京广外医院的脑卒中患者48例,分为对照组22例和治疗组26例,设计脑卒中社区康复简易技术训练,结合常规治疗,治疗组患者进行集体训练,对照组患者仅进行常规治疗。
6) cluster training
群集训练
补充资料:训练集
分子式:
CAS号:
性质:又称训练集。在人工神经网络法中,用一定容量的样本值来训练网络,求得各神经元之间的连接权重,以建立网络模型,用来训练网络的样本值,称为学习样本。
CAS号:
性质:又称训练集。在人工神经网络法中,用一定容量的样本值来训练网络,求得各神经元之间的连接权重,以建立网络模型,用来训练网络的样本值,称为学习样本。
说明:补充资料仅用于学习参考,请勿用于其它任何用途。
参考词条