The qualnet kernel and api structure were carefully designed to facilitate development , making it easy for a user to develop custom parallel models 結(jié)構(gòu)經(jīng)過認(rèn)真的設(shè)計(jì),以便于開發(fā),對(duì)于開發(fā)定制的并行模型的用戶來說非常簡(jiǎn)單。
At last , this paper uses the rbf neural network that was optimized by the mentioned parallel model to predict the value of some nonlinear functions and the close of several stocks 最后,本文應(yīng)用上述并行模型優(yōu)化的rbf神經(jīng)網(wǎng)絡(luò)對(duì)非線性函數(shù)值以及證券個(gè)股收盤價(jià)進(jìn)行預(yù)測(cè)。
Basing on the muti - threads parallel model on shered memory , we make some research and test of program parallelization on source code level and intermediate representation level 根據(jù)基于共享內(nèi)存方式的多線程并行模型,對(duì)于在源代碼級(jí)和中間代碼級(jí)進(jìn)行程序并行化做了一些研究和實(shí)驗(yàn)。