Yuantao Feng commited on
Commit
2ef6bc9
·
1 Parent(s): 3af1dea

Add hardware GPU CUDA: Jetson Nano (#8)

Browse files

* update benchmark results on Jetson Nano

Files changed (1) hide show
  1. README.md +9 -16
README.md CHANGED
@@ -11,30 +11,23 @@ Guidelines:
11
  Hardware Setup:
12
  - `CPU x86_64`: INTEL CPU i7-5930K @ 3.50GHz, 6 cores, 12 threads.
13
  - `CPU ARM`: Raspberry 4B, BCM2711B0 @ 1.5GHz (Cortex A-72), 4 cores, 4 threads.
14
- <!--
15
  - `GPU CUDA`: NVIDIA Jetson Nano B01, 128-core Maxwell, Quad-core ARM A57 @ 1.43 GHz.
16
- -->
17
 
18
  ***Important Notes***:
19
  - The time data that shown on the following table presents the time elapsed from preprocess (resize is excluded), to a forward pass of a network, and postprocess to get final results.
20
  - The time data that shown on the following table is the median of 10 runs. Different metrics may be applied to some specific models.
21
  - View [benchmark/config](./benchmark/config) for more details on benchmarking different models.
22
 
23
- <!--
24
- | Model | Input Size | CPU x86_64 (ms) | CPU ARM (ms) | GPU CUDA (ms) |
25
  |-------|------------|-----------------|--------------|---------------|
26
- | [YuNet](./models/face_detection_yunet) | 160x120 | 2.17 | 8.87 | 14.95 |
27
- | [DB](./models/text_detection_db) | 640x480 | 148.65 | 2759.88 | 218.25 |
28
- | [CRNN](./models/text_recognition_crnn) | 100x32 | 23.23 | 235.87 | 195.20 |
29
- -->
30
- | Model | Input Size | CPU x86_64 (ms) | CPU ARM (ms) |
31
- |-------|------------|-----------------|--------------|
32
- | [YuNet](./models/face_detection_yunet) | 160x120 | 1.45 | 6.22 |
33
- | [DB](./models/text_detection_db) | 640x480 | 137.38 | 2780.78 |
34
- | [CRNN](./models/text_recognition_crnn) | 100x32 | 50.21 | 234.32 |
35
- | [SFace](./models/face_recognition_sface) | 112x112 | 8.65 | 99.20 |
36
- | [PP-ResNet](./models/image_classification_ppresnet) | 224x224 | 56.05 | 602.58
37
- | [PP-HumanSeg](./models/human_segmentation_pphumanseg) | 192x192 | 19.92 | 105.32 |
38
 
39
  ## License
40
 
 
11
  Hardware Setup:
12
  - `CPU x86_64`: INTEL CPU i7-5930K @ 3.50GHz, 6 cores, 12 threads.
13
  - `CPU ARM`: Raspberry 4B, BCM2711B0 @ 1.5GHz (Cortex A-72), 4 cores, 4 threads.
 
14
  - `GPU CUDA`: NVIDIA Jetson Nano B01, 128-core Maxwell, Quad-core ARM A57 @ 1.43 GHz.
 
15
 
16
  ***Important Notes***:
17
  - The time data that shown on the following table presents the time elapsed from preprocess (resize is excluded), to a forward pass of a network, and postprocess to get final results.
18
  - The time data that shown on the following table is the median of 10 runs. Different metrics may be applied to some specific models.
19
  - View [benchmark/config](./benchmark/config) for more details on benchmarking different models.
20
 
21
+ | Model | Input Size | CPU x86_64 (ms) | CPU ARM (ms) | GPU CUDA (ms)* |
 
22
  |-------|------------|-----------------|--------------|---------------|
23
+ | [YuNet](./models/face_detection_yunet) | 160x120 | 1.45 | 6.22 | 12.18 |
24
+ | [DB](./models/text_detection_db) | 640x480 | 137.38 | 2780.78 | 220.52 |
25
+ | [CRNN](./models/text_recognition_crnn) | 100x32 | 50.21 | 234.32 | 196.15 |
26
+ | [SFace](./models/face_recognition_sface) | 112x112 | 8.65 | 99.20 | 24.88 |
27
+ | [PP-ResNet](./models/image_classification_ppresnet) | 224x224 | 56.05 | 602.58 | 98.64 |
28
+ | [PP-HumanSeg](./models/human_segmentation_pphumanseg) | 192x192 | 19.92 | 105.32 | 67.97 |
29
+
30
+ *: Batch size is 1.
 
 
 
 
31
 
32
  ## License
33