Research Article
Accelerating Relevance-Vector-Machine-Based Classification of Hyperspectral Image with Parallel Computing
Table 6
The cost (seconds) and speedup ratio (in parenthesis) of pRVM-MatOp on multicore platforms.
| Data set | Data size | Two-core server | Four-core workstation | | 1 Core | 2 Cores | 1 Core | 4 Cores | |
| DS-A | 848 (20%) | 21.5 | 16.5 (1.30) | 22.7 | 15.2 (1.49) | | DS-B | 1697 (40%) | 147.4 | 113.9 (1.29) | 152.8 | 86.3 (1.77) | | DS-C | 2546 (60%) | 409.7 | 312.3 (1.31) | 427.4 | 207.5 (2.06) | | DS-D | 3395 (80%) | 832.5 | 634.1 (1.31) | 868.5 | 366.5 (2.37) | | DS-E | 4244 (100%) | 1354.9 | 1028.3 (1.32) | 1413.5 | 559.3 (2.53) | |
|
|