I have been getting the #Target has been reduced to 25 due to too many failures or duplications error, while trying to cross compile (retinanet model), for a jetson nano target.
If my target is cuda, no such error occurs and auto scheduling can be performed smoothly.
Has anyone here been able to cross compile for a jetson nano target using auto scheduler without any issue? I am doubting if it is even possible as long as the target is jetson nano.
Generate Sketches #s: 2
Sample Iter: 5 #Pop: 10 #Target: 50 fail_ct: 10230 Time elapsed: 2.67
#Target has been reduced to 25 due to too many failures or duplications
Sample Iter: 10 #Pop: 10 #Target: 25 fail_ct: 20470 Time elapsed: 6.61
#Target has been reduced to 12 due to too many failures or duplications
Sample Iter: 15 #Pop: 10 #Target: 12 fail_ct: 30710 Time elapsed: 9.62
#Target has been reduced to 6 due to too many failures or duplications
Sample Initial Population #s: 10 fail_ct: 32758 Time elapsed: 10.17
GA Iter: 0 Max score: 0.8786 Min score: 0.1769 #Pop: 8 #M+: 0 #M-: 0
GA Iter: 4 Max score: 0.9999 Min score: 0.9991 #Pop: 8 #M+: 1396 #M-: 0
EvolutionarySearch #s: 8 Time elapsed: 1.81
----------------------------------------------------------------------
------------------------------ [ Measure ]
----------------------------------------------------------------------
Get 4 programs to measure:
....
*E*E*E*ETime elapsed for measurement: 3.16 s
----------------------------------------------------------------------
------------------------------ [ Train cost model ]
----------------------------------------------------------------------
Time elapsed for training: 0.03 s
----------------------------------------------------------------------
------------------------------ [ Task Scheduler ]
----------------------------------------------------------------------
| ID | Task Description | Latency (ms) | Speed (GFLOPS) | Trials |
-----------------------------------------------------------------------------------------------------------------
| 0 | vm_mod_fused_nn_conv2d_add_add_1 | - | - | 4 |
| 1 | vm_mod_fused_max | - | - | 4 |
| 2 | vm_mod_fused_nn_conv2d_add_add_nn_relu_2 | - | - | 0 |
| 3 | vm_mod_fused_nn_conv2d_add_2 | - | - | 0 |
| 4 | vm_mod_fused_nn_conv2d_add_nn_relu_11 | - | - | 0 |
| 5 | vm_mod_fused_nn_conv2d_add_nn_relu_8 | - | - | 0 |
| 6 | vm_mod_fused_nn_max_pool2d | - | - | 0 |
| 7 | vm_mod_fused_nn_conv2d_add_13 | - | - | 0 |
| 8 | vm_mod_fused_nn_conv2d_add_11 | - | - | 0 |
| 9 | vm_mod_fused_nn_conv2d_add_add | - | - | 0 |
| 10 | vm_mod_fused_nn_conv2d_add_nn_relu_7 | - | - | 0 |
| 11 | vm_mod_fused_nn_conv2d_add | - | - | 0 |
| 12 | vm_mod_fused_nn_conv2d_add_1 | - | - | 0 |
| 13 | vm_mod_fused_max_2 | - | - | 0 |
| 14 | vm_mod_fused_max_4 | - | - | 0 |
| 15 | vm_mod_fused_nn_conv2d_add_16 | - | - | 0 |
| 16 | vm_mod_fused_nn_conv2d_add_8 | - | - | 0 |
| 17 | vm_mod_fused_nn_conv2d_add_12 | - | - | 0 |
| 18 | vm_mod_fused_max_3 | - | - | 0 |
| 19 | vm_mod_fused_prod_2 | - | - | 0 |
| 20 | vm_mod_fused_nn_conv2d_add_4 | - | - | 0 |
| 21 | vm_mod_fused_nn_conv2d_add_3 | - | - | 0 |
| 22 | vm_mod_fused_nn_conv2d_add_nn_relu_5 | - | - | 0 |
| 23 | vm_mod_fused_nn_conv2d_add_nn_relu_6 | - | - | 0 |
| 24 | vm_mod_fused_nn_conv2d_add_nn_relu_4 | - | - | 0 |
| 25 | vm_mod_fused_nn_conv2d_add_add_nn_relu | - | - | 0 |
| 26 | vm_mod_fused_nn_conv2d_add_nn_relu_3 | - | - | 0 |
| 27 | vm_mod_fused_nn_conv2d_add_nn_relu_10 | - | - | 0 |
| 28 | vm_mod_fused_nn_conv2d_add_5 | - | - | 0 |
| 29 | vm_mod_fused_nn_conv2d_add_add_nn_relu_3 | - | - | 0 |
| 30 | vm_mod_fused_nn_conv2d_add_14 | - | - | 0 |
| 31 | vm_mod_fused_nn_conv2d_add_add_nn_relu_1 | - | - | 0 |
| 32 | vm_mod_fused_nn_conv2d_add_nn_relu_9 | - | - | 0 |
| 33 | vm_mod_fused_max_5 | - | - | 0 |
| 34 | vm_mod_fused_nn_conv2d_add_9 | - | - | 0 |
| 35 | vm_mod_fused_nn_conv2d_add_nn_relu_2 | - | - | 0 |
| 36 | vm_mod_fused_prod_1 | - | - | 0 |
| 37 | vm_mod_fused_nn_conv2d_add_6 | - | - | 0 |
| 38 | vm_mod_fused_nn_conv2d_add_nn_relu | - | - | 0 |
| 39 | vm_mod_fused_prod | - | - | 0 |
| 40 | vm_mod_fused_nn_conv2d_add_10 | - | - | 0 |
| 41 | vm_mod_fused_nn_conv2d_add_17 | - | - | 0 |
| 42 | vm_mod_fused_nn_conv2d_add_15 | - | - | 0 |
| 43 | vm_mod_fused_nn_conv2d_add_18 | - | - | 0 |
| 44 | vm_mod_fused_nn_conv2d_add_7 | - | - | 0 |
| 45 | vm_mod_fused_nn_conv2d_add_nn_relu_1 | - | - | 0 |
| 46 | vm_mod_fused_max_1 | - | - | 0 |
-----------------------------------------------------------------------------------------------------------------
Estimated total latency: - ms Trials: 8 Used time : 59 s Next ID: 2