Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- PS C:\DarknetOpenCV\DarkPlate> darknet detector -map -dont_show train nn/own_darkmark_dataset.data nn/own_darkmark_dataset.cfg
- CUDA-version: 11000 (11070), cuDNN: 8.0.5, GPU count: 1
- OpenCV version: 4.6.0
- Prepare additional network for mAP calculation...
- 0 : compute_capability = 610, cudnn_half = 0, GPU: NVIDIA GeForce GTX 1070
- net.optimized_memory = 0
- mini_batch = 1, batch = 4, time_steps = 1, train = 0
- layer filters size/strd(dil) input output
- 0 Create CUDA-stream - 0
- Create cudnn-handle 0
- conv 32 3 x 3/ 2 1152 x 864 x 3 -> 576 x 432 x 32 0.430 BF
- 1 conv 64 3 x 3/ 2 576 x 432 x 32 -> 288 x 216 x 64 2.293 BF
- 2 conv 64 3 x 3/ 1 288 x 216 x 64 -> 288 x 216 x 64 4.586 BF
- 3 route 2 1/2 -> 288 x 216 x 32
- 4 conv 32 3 x 3/ 1 288 x 216 x 32 -> 288 x 216 x 32 1.147 BF
- 5 conv 32 3 x 3/ 1 288 x 216 x 32 -> 288 x 216 x 32 1.147 BF
- 6 route 5 4 -> 288 x 216 x 64
- 7 conv 64 1 x 1/ 1 288 x 216 x 64 -> 288 x 216 x 64 0.510 BF
- 8 route 2 7 -> 288 x 216 x 128
- 9 max 2x 2/ 2 288 x 216 x 128 -> 144 x 108 x 128 0.008 BF
- 10 conv 128 3 x 3/ 1 144 x 108 x 128 -> 144 x 108 x 128 4.586 BF
- 11 route 10 1/2 -> 144 x 108 x 64
- 12 conv 64 3 x 3/ 1 144 x 108 x 64 -> 144 x 108 x 64 1.147 BF
- 13 conv 64 3 x 3/ 1 144 x 108 x 64 -> 144 x 108 x 64 1.147 BF
- 14 route 13 12 -> 144 x 108 x 128
- 15 conv 128 1 x 1/ 1 144 x 108 x 128 -> 144 x 108 x 128 0.510 BF
- 16 route 10 15 -> 144 x 108 x 256
- 17 max 2x 2/ 2 144 x 108 x 256 -> 72 x 54 x 256 0.004 BF
- 18 conv 256 3 x 3/ 1 72 x 54 x 256 -> 72 x 54 x 256 4.586 BF
- 19 route 18 1/2 -> 72 x 54 x 128
- 20 conv 128 3 x 3/ 1 72 x 54 x 128 -> 72 x 54 x 128 1.147 BF
- 21 conv 128 3 x 3/ 1 72 x 54 x 128 -> 72 x 54 x 128 1.147 BF
- 22 route 21 20 -> 72 x 54 x 256
- 23 conv 256 1 x 1/ 1 72 x 54 x 256 -> 72 x 54 x 256 0.510 BF
- 24 route 18 23 -> 72 x 54 x 512
- 25 max 2x 2/ 2 72 x 54 x 512 -> 36 x 27 x 512 0.002 BF
- 26 conv 512 3 x 3/ 1 36 x 27 x 512 -> 36 x 27 x 512 4.586 BF
- 27 conv 256 1 x 1/ 1 36 x 27 x 512 -> 36 x 27 x 256 0.255 BF
- 28 conv 512 3 x 3/ 1 36 x 27 x 256 -> 36 x 27 x 512 2.293 BF
- 29 conv 126 1 x 1/ 1 36 x 27 x 512 -> 36 x 27 x 126 0.125 BF
- 30 yolo
- [yolo] params: iou loss: ciou (4), iou_norm: 0.07, obj_norm: 1.00, cls_norm: 1.00, delta_norm: 1.00, scale_x_y: 1.05
- nms_kind: greedynms (1), beta = 0.600000
- 31 route 27 -> 36 x 27 x 256
- 32 conv 128 1 x 1/ 1 36 x 27 x 256 -> 36 x 27 x 128 0.064 BF
- 33 upsample 2x 36 x 27 x 128 -> 72 x 54 x 128
- 34 route 33 23 -> 72 x 54 x 384
- 35 conv 256 3 x 3/ 1 72 x 54 x 384 -> 72 x 54 x 256 6.880 BF
- 36 conv 126 1 x 1/ 1 72 x 54 x 256 -> 72 x 54 x 126 0.251 BF
- 37 yolo
- [yolo] params: iou loss: ciou (4), iou_norm: 0.07, obj_norm: 1.00, cls_norm: 1.00, delta_norm: 1.00, scale_x_y: 1.05
- nms_kind: greedynms (1), beta = 0.600000
- Total BFLOPS 39.359
- avg_outputs = 1751134
- Allocate additional workspace_size = 9.45 MB
- own_darkmark_dataset
- 0 : compute_capability = 610, cudnn_half = 0, GPU: NVIDIA GeForce GTX 1070
- net.optimized_memory = 0
- mini_batch = 16, batch = 64, time_steps = 1, train = 1
- layer filters size/strd(dil) input output
- 0 conv 32 3 x 3/ 2 1152 x 864 x 3 -> 576 x 432 x 32 0.430 BF
- 1 conv 64 3 x 3/ 2 576 x 432 x 32 -> 288 x 216 x 64 2.293 BF
- 2 conv 64 3 x 3/ 1 288 x 216 x 64 -> 288 x 216 x 64 4.586 BF
- 3 route 2 1/2 -> 288 x 216 x 32
- 4 conv 32 3 x 3/ 1 288 x 216 x 32 -> 288 x 216 x 32 1.147 BF
- 5 conv 32 3 x 3/ 1 288 x 216 x 32 -> 288 x 216 x 32 1.147 BF
- 6 route 5 4 -> 288 x 216 x 64
- 7 conv 64 1 x 1/ 1 288 x 216 x 64 -> 288 x 216 x 64 0.510 BF
- 8 route 2 7 -> 288 x 216 x 128
- 9 max 2x 2/ 2 288 x 216 x 128 -> 144 x 108 x 128 0.008 BF
- 10 conv 128 3 x 3/ 1 144 x 108 x 128 -> 144 x 108 x 128 4.586 BF
- 11 route 10 1/2 -> 144 x 108 x 64
- 12 Error: cuDNN isn't found FWD algo for convolution.
Advertisement
Add Comment
Please, Sign In to add comment
