A mixed hardware/software SOFM training system
Fuente:
Redalyc
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Artículo científico |
| Lingua: | en |
| Pubblicazione: |
Instituto Politécnico Nacional
2008
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1876432192340492288 |
|---|---|
| author | Agustín Ramírez Agundis |
| author_facet | Agustín Ramírez Agundis |
| contents | A mixed hardware/software SOFM training system Agustín Ramírez Agundis Rafael Gadea Girones Ricardo Colom Palero Javier Díaz Carmona Computación Field Pro Mixed Hardware Neural coprocessor grammable Gate Array Software Implementation This paper describes the design of a training system for a Self-Organizing Feature Map (SOFM). The system design aims two goals. The first is to reduce the training processing time by exploiting the inherent neural networks (NNs) parallelism through the SOFM hardware implementation. The second goal is to provide versatility to the training process by means of pre- and post processing of input and output data using Matlab-Simulink, which is also used as the software platform. The sys-tem uses as a coprocessor an FPGA based board connected via PCI bus at the host PC. To illu-strate the system functionality we developed an application to analyze the effects over the map of scattering size in randomly generated weight initial values. When compared with the software ap-proach for the same application, our system reduces the training time in 89%. 2008 artículo científico 1405-5546 https://www.redalyc.org/articulo.oa?id=61511404 en http://www.redalyc.org/revista.oa?id=615 Computación y Sistemas application/pdf Instituto Politécnico Nacional Computación y Sistemas (México) Num.4 Vol.11 |
| format | Artículo científico |
| id | redalyc_61511404 |
| institution | Redalyc |
| language | en |
| publishDate | 2008 |
| publisher | Instituto Politécnico Nacional |
| spellingShingle | A mixed hardware/software SOFM training system Agustín Ramírez Agundis Computación Field Pro Mixed Hardware Neural coprocessor grammable Gate Array Software Implementation A mixed hardware/software SOFM training system Agustín Ramírez Agundis Rafael Gadea Girones Ricardo Colom Palero Javier Díaz Carmona Computación Field Pro Mixed Hardware Neural coprocessor grammable Gate Array Software Implementation This paper describes the design of a training system for a Self-Organizing Feature Map (SOFM). The system design aims two goals. The first is to reduce the training processing time by exploiting the inherent neural networks (NNs) parallelism through the SOFM hardware implementation. The second goal is to provide versatility to the training process by means of pre- and post processing of input and output data using Matlab-Simulink, which is also used as the software platform. The sys-tem uses as a coprocessor an FPGA based board connected via PCI bus at the host PC. To illu-strate the system functionality we developed an application to analyze the effects over the map of scattering size in randomly generated weight initial values. When compared with the software ap-proach for the same application, our system reduces the training time in 89%. 2008 artículo científico 1405-5546 https://www.redalyc.org/articulo.oa?id=61511404 en http://www.redalyc.org/revista.oa?id=615 Computación y Sistemas application/pdf Instituto Politécnico Nacional Computación y Sistemas (México) Num.4 Vol.11 |
| title | A mixed hardware/software SOFM training system |
| topic | Computación Field Pro Mixed Hardware Neural coprocessor grammable Gate Array Software Implementation |
| url | https://www.redalyc.org/articulo.oa?id=61511404 |