. | . | |||||
---|---|---|---|---|---|---|
. | 1.0 . | 0.5 . | 0.25 . | 0.15 . | 0.1 . | 0.01 . |
1-primitive prediction (%) | 81.25 | 90.63 | 100 | 93.75 | 90.63 | 81.25 |
2-primitive prediction (%) | 62.50 | 78.13 | 84.38 | 68.75 | 59.38 | 37.50 |
3-primitive prediction (%) | 40.63 | 53.13 | 59.38 | 37.50 | 21.88 | 9.38 |
. | . | |||||
---|---|---|---|---|---|---|
. | 1.0 . | 0.5 . | 0.25 . | 0.15 . | 0.1 . | 0.01 . |
1-primitive prediction (%) | 81.25 | 90.63 | 100 | 93.75 | 90.63 | 81.25 |
2-primitive prediction (%) | 62.50 | 78.13 | 84.38 | 68.75 | 59.38 | 37.50 |
3-primitive prediction (%) | 40.63 | 53.13 | 59.38 | 37.50 | 21.88 | 9.38 |
Notes: Intermediate values of the meta-prior reproduce best the sequence of primitives. The table shows the percent accuracy for one, two, and three primitive prediction for models trained with different values for . The network with the best accuracy is shown in bold numbers.