- About
-
The Global ETD Search service is a free service for researchers
to find electronic theses and dissertations. This service is
provided by the
Networked Digital Library of Theses and Dissertations.
Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
Search results
Showing 1 to 2 of 2 (0.117 seconds)Spelling suggestions: "subject:"högpresterande näringsberäkningar"" "subject:"högpresterande sättningsberäkningar""
| 1 |
Leveraging Posits for the Conjugate Gradient Linear Solver on an Application-Level RISC-V CoreMallasén Quintana, David January 2022 (has links)
Emerging floating-point arithmetics provide a way to optimize the execution of computationally-intensive algorithms. This is the case with scientific computational kernels such as the Conjugate Gradient (CG) linear solver. Exploring new arithmetics is of paramount importance to maximize the accuracy and timing performance of these algorithms. In this thesis, I have studied the use of the novel posit arithmetic in hardware to improve the accuracy of the CG method. In particular, on PERCIVAL, an application-level RISC-V core with support for posits and quire. The open RISC-V architecture supplies a flexible platform for the exploration of new computer architecture studies. Previous works have tackled the use of posits in the high-performance computing and machine learning fields, amongst others. However, until recently, the lack of hardware support has been a significant barrier to their scalability. The key results from this thesis show that posits are a promising alternative when solving 1D and 2D Poisson equations using the CG linear solver. Notably, this novel arithmetic can execute as fast as IEEE 754 floating-point numbers on specialized hardware, and provide up to 2 orders of magnitude higher accuracy. This accuracy improvement spans both the error of the output values of the algorithms and the value of the final residual in the CG iterative method. Furthermore, the use of the quire accumulator register in the computation of dot-products in posit arithmetic significantly boosts the accuracy of the outputs. Since 32-bit posits perform practically as fast as 32-bit floats, and thus faster than 64-bit floats, they present an intermediate solution between single- and double-precision arithmetic. This paves the way for the deployment of high-efficiency solutions that make intensive use of floating-point operations. / Ny kommande flyttalsaritmetik ger ett sätt att optimera exekveringen av beräkningsintensiva algoritmer. Detta är fallet med vetenskapliga beräkningskärnor som den Conjugate Gradient (CG) metoden kräver. Att utforska ny aritmetik är av största vikt för att minska energikostnaderna för dessa algoritmer. I detta examensarbete har jag studerat användningen av den nya positaritmetiken i hårdvara för att förbättra noggrannheten i CG-metoden. I synnerhet på PERCIVAL, en RISC-V-kärna på applikationsnivå med stöd för posits och quire. Den öppna RISC-V-arkitekturen tillhandahåller en flexibel plattform för utforskning av nya dator arkitekturstudier. Tidigare arbeten har tagit itu med användningen av positurer inom områdena högpresterande datorer och maskininlärning, bland annat. Men fram till nyligen har bristen på hårdvarustöd varit ett betydande hinder för deras skalbarhet. Nyckelresultaten från denna avhandling visar att posits är ett lovande alternativ när man löser 1D och 2D Poisson-ekvationer med den linjära CG-lösaren. Noterbart kan denna nya aritmetik köra så snabbt som IEEE 754 flyttal på specialiserad hårdvara och ge upp till två storleksordningar högre noggrannhet. Denna noggrannhetsförbättring sträcker sig över både felet i algoritmernas utvärden och värdet på den slutliga residualen i den iterativa CG-metoden. Dessutom ökar användningen av quire-ackumulatorregistret vid beräkning av punktprodukter i positaritmetik avsevärt noggrannheten hos utsignalerna. Eftersom 32-bitars posits presterar praktiskt taget lika snabbt som 32-bitars flöten, och därmed snabbare än 64-bitars flöten, presenterar de en mellanlösning mellan enkel-och dubbelprecisionsaritmetik. Detta banar väg för utbyggnaden av högeffektiva lösningar som intensivt utnyttjar flyttalsoperationer.
|
| 2 |
Revision of an artificial neural network enabling industrial sortingMalmgren, Henrik January 2019 (has links)
Convolutional artificial neural networks can be applied for image-based object classification to inform automated actions, such as handling of objects on a production line. The present thesis describes theoretical background for creating a classifier and explores the effects of introducing a set of relatively recent techniques to an existing ensemble of classifiers in use for an industrial sorting system.The findings indicate that it's important to use spatial variety dropout regularization for high resolution image inputs, and use an optimizer configuration with good convergence properties. The findings also demonstrate examples of ensemble classifiers being effectively consolidated into unified models using the distillation technique. An analogue arrangement with optimization against multiple output targets, incorporating additional information, showed accuracy gains comparable to ensembling. For use of the classifier on test data with statistics different than those of the dataset, results indicate that augmentation of the input data during classifier creation helps performance, but would, in the current case, likely need to be guided by information about the distribution shift to have sufficiently positive impact to enable a practical application. I suggest, for future development, updated architectures, automated hyperparameter search and leveraging the bountiful unlabeled data potentially available from production lines.
|
Page generated in 0.1226 seconds