Skip to main content
. 2009 Sep 11;9(9):7308–7319. doi: 10.3390/s90907308

Table 1.

Results of the CMAC Neural Network Algorithm for the odor data.

Q W β Desired Error Init Time (s) Learning Time (s) Test Time (s) Step
4 2 0.4 0.1 1,390 0.438 0.031 4
4 2 0.4 0.01 1,375 0.578 0.031 5
4 2 0.4 0.001 1,391 0.640 0.031 6
4 2 0.4 0.0001 1,359 1,282 0.015 12
4 2 0.4 0.00001 1,344 1,250 0.032 18
4 2 0.4 0.000001 1,375 1,860 0.031 23

4 2 0.6 0.1 1,359 0.453 0.047 4
4 2 0.6 0.01 1,375 0.562 0.016 5
4 2 0.6 0.001 1,391 0.640 0.031 6
4 2 0.6 0.0001 1,390 1,266 0.015 12
4 2 0.6 0.00001 1,391 1,953 0.016 18
4 2 0.6 0.000001 1,391 2,438 0.015 23

4 2 0.8 0.1 1,391 0.469 0.031 4
4 2 0.8 0.01 1,375 0.578 0.016 5
4 2 0.8 0.001 1,390 0.657 0.015 6
4 2 0.8 0.0001 1,360 1,094 0.031 10
4 2 0.8 0.00001 1,407 1,656 0.031 16
4 2 0.8 0.000001 1,406 2,250 0.032 21