Skip to content

Commit

Permalink
Update README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
@andyluo7
andyluo7 authored Jul 23, 2020
1 parent 9772d56 commit 6a699e3
Showing 1 changed file with 35 additions and 35 deletions.
70 changes: 35 additions & 35 deletions AI-Model-Zoo/README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -355,30 +355,30 @@ The following table lists the performance number including end-to-end throughput

| No\. | Model | Name | E2E latency \(ms\) Thread num =1 | E2E throughput \-fps\(Single Thread\) | E2E throughput \-fps\(Multi Thread\) |
| ---- | :------------------------- | :------------------------------------------- | -------------------------------- | ------------------------------------- | ------------------------------------ |
| 1 | resnet50 | cf_resnet50_imagenet_224_224_7.7G | 13.99 | 71.5 | 153.0 |
| 2 | resnet18 | cf_resnet18_imagenet_224_224_3.65G | 5.74 | 174.2 | 442.5 |
| 3 | Inception_v1 | cf_inceptionv1_imagenet_224_224_3.16G | 6.00 | 166.7 | 414.0 |
| 4 | Inception_v2 | cf_inceptionv2_imagenet_224_224_4G | 7.31 | 136.8 | 318.9 |
| 5 | Inception_v3 | cf_inceptionv3_imagenet_299_299_11.4G | 17.76 | 56.3 | 128.7 |
| 6 | Inception_v4 | cf_inceptionv4_imagenet_299_299_24.5G | 35.45 | 28.2 | 66.2 |
| 7 | Mobilenet_v2 | cf_mobilenetv2_imagenet_224_224_0.59G | 4.79 | 208.6 | 550.9 |
| 8 | SqueezeNet | cf_squeeze_imagenet_227_227_0.76G | 3.76 | 265.8 | 1013.7 |
| 1 | resnet50 | cf_resnet50_imagenet_224_224_7.7G | 13.60 | 73.5 | 152.7 |
| 2 | resnet18 | cf_resnet18_imagenet_224_224_3.65G | 5.35 | 186.9 | 441.6 |
| 3 | Inception_v1 | cf_inceptionv1_imagenet_224_224_3.16G | 5.62 | 178.0 | 411.7 |
| 4 | Inception_v2 | cf_inceptionv2_imagenet_224_224_4G | 6.92 | 144.4 | 317.3 |
| 5 | Inception_v3 | cf_inceptionv3_imagenet_299_299_11.4G | 17.38 | 57.5 | 128.1 |
| 6 | Inception_v4 | cf_inceptionv4_imagenet_299_299_24.5G | 35.06 | 28.5 | 66.2 |
| 7 | Mobilenet_v2 | cf_mobilenetv2_imagenet_224_224_0.59G | 4.41 | 226.8 | 548.0 |
| 8 | SqueezeNet | cf_squeeze_imagenet_227_227_0.76G | 3.76 | 265.8 | 1012.3 |
| 9 | ssd_pedestrian_pruned_0_97 | cf_ssdpedestrian_coco_360_640_0.97_5.9G | 13.10 | 76.3 | 282.6 |
| 10 | refinedet\_baseline | cf_refinedet_coco_480_360_123G_1.2 | 120.96 | 8.3 | 24.4 |
| 10 | refinedet\_baseline | cf_refinedet_coco_480_360_123G_1.2 | 120.96 | 8.3 | 24.4 |
| 11 | refinedet_pruned_0_8 | cf_refinedet_coco_360_480_0.8_25G | 31.58 | 31.7 | 101.3 |
| 12 | refinedet_pruned_0_92 | cf_refinedet_coco_360_480_0.92_10.10G | 16.68 | 59.9 | 196.8 |
| 13 | refinedet_pruned_0_96 | cf_refinedet_coco_360_480_0.96_5.08G | 12.07 | 82.9 | 276.2 |
| 14 | ssd_adas_pruned_0_95 | cf_ssdadas_bdd_360_480_0.95_6.3G | 12.05 | 82.9 | 279.7 |
| 15 | ssd_traffic_pruned_0_9 | cf_ssdtraffic_360_480_0.9_11.6G | 18.33 | 54.5 | 201.5 |
| 16 | VPGnet_pruned_0_99 | cf_VPGnet_caltechlane_480_640_0.99_2.5G | 9.56 | 104.5 | 381.4 |
| 16 | VPGnet_pruned_0_99 | cf_VPGnet_caltechlane_480_640_0.99_2.5G | 9.56 | 104.5 | 381.4 |
| 17 | ssd_mobilenet_v2 | cf_ssdmobilenetv2_bdd_360_480_6.57G | 26.01 | 38.4 | 114.4 |
| 18 | FPN | cf_fpn_cityscapes_256_512_8.9G | 16.74 | 59.7 | 175.5 |
| 19 | SP_net | cf_SPnet_aichallenger_224_128_0.54G | 2.62 | 381.6 | 1317.4 |
| 20 | Openpose_pruned_0_3 | cf_openpose_aichallenger_368_368_0.3_189.7G | 285.58 | 3.5 | 15.1 |
| 21 | densebox_320_320 | cf_densebox_wider_320_320_0.49G | 2.56 | 390.0 | 1172.3 |
| 22 | densebox_640_360 | cf_densebox_wider_360_640_1.11G | 4.99 | 200.4 | 588.7 |
| 23 | face_landmark | cf_landmark_celeba_96_72_0.14G | 1.18 | 849.4 | 1382.7 |
| 24 | reid | cf_reid_market1501_160_80_0.95G | 2.74 | 364.2 | 665.6 |
| 19 | SP_net | cf_SPnet_aichallenger_224_128_0.54G | 2.62 | 381.6 | 1317.4 |
| 20 | Openpose_pruned_0_3 | cf_openpose_aichallenger_368_368_0.3_189.7G | 285.58 | 3.5 | 15.1 |
| 21 | densebox_320_320 | cf_densebox_wider_320_320_0.49G | 2.56 | 390.0 | 1172.3 |
| 22 | densebox_640_360 | cf_densebox_wider_360_640_1.11G | 4.99 | 200.4 | 588.7 |
| 23 | face_landmark | cf_landmark_celeba_96_72_0.14G | 1.18 | 849.4 | 1382.7 |
| 24 | reid | cf_reid_market1501_160_80_0.95G | 2.74 | 364.2 | 665.6 |
| 25 | multi_task | cf_multitask_bdd_288_512_14.8G | 28.19 | 35.5 | 127.7 |
| 26 | yolov3_bdd | dk_yolov3_bdd_288_512_53.7G | 77.10 | 13.0 | 34.3 |
| 27 | yolov3_adas_pruned_0_9 | dk_yolov3_cityscapes_256_512_0.9_5.46G | 11.89 | 84.1 | 229.7 |
Expand All @@ -387,30 +387,30 @@ The following table lists the performance number including end-to-end throughput
| 30 | yolov2_voc_pruned_0_66 | dk_yolov2_voc_448_448_0.66_11.56G | 15.82 | 63.2 | 185.9 |
| 31 | yolov2_voc_pruned_0_71 | dk_yolov2_voc_448_448_0.71_9.86G | 13.73 | 72.8 | 214.8 |
| 32 | yolov2_voc_pruned_0_77 | dk_yolov2_voc_448_448_0.77_7.82G | 11.73 | 85.2 | 258.7 |
| 33 | ResNet20-face | cf_facerec-resnet20_112_96_3.5G | 5.98 | 167.1 | 320.6 |
| 33 | ResNet20-face | cf_facerec-resnet20_112_96_3.5G | 5.98 | 167.1 | 320.6 |
| 34 | ResNet64-face | cf_facerec-resnet64_112_96_11G | 13.69 | 73.0 | 173.0 |
| 35 | FPN_Res18_segmentation | cf_FPN-resnet18_EDD_320_320_45.3G | 81.86 | 12.2 | 40.3 |
| 36 | plate detection | cf_plate-detection_320_320_0.49G | 2.00 | 500.0 | 1792.2 |
| 37 | plate recognition | cf_plate-recognition_96_288_1.75G | 8.82 | 113.4 | 383.2 |
| 38 | Inception_resnet_v2 | tf_inceptionresnetv2_imagenet_299_299_26.35G | 43.58 | 22.9 | 48.8 |
| 39 | Inception_v1 | tf_inceptionv1_imagenet_224_224_3G | 5.82 | 171.9 | 426.4 |
| 40 | Inception_v3 | tf_inceptionv3_imagenet_299_299_11.45G | 17.83 | 56.1 | 127.3 |
| 41 | Inception_v4 | tf_inceptionv4_imagenet_299_299_24.55G | 35.47 | 28.2 | 66.3 |
| 42 | Mobilenet_v1 | tf_mobilenetv1_0.25_imagenet_128_128_27.15M | 1.24 | 805.8 | 2190.7 |
| 43 | Mobilenet_v1 | tf_mobilenetv1_0.5_imagenet_160_160_150.07M | 1.80 | 555.8 | 1922.6 |
| 44 | Mobilenet_v1 | tf_mobilenetv1_1.0_imagenet_224_224_1.14G | 3.90 | 256.1 | 758.4 |
| 45 | Mobilenet_v2 | tf_mobilenetv2_1.0_imagenet_224_224_0.59G | 4.72 | 211.8 | 571.2 |
| 46 | Mobilenet_v2 | tf_mobilenetv2_1.4_imagenet_224_224_1.16G | 6.36 | 157.2 | 395.0 |
| 47 | resnet_v1_50 | tf_resnetv1_50_imagenet_224_224_6.97G | 13.02 | 76.7 | 162.2 |
| 48 | resnet_v1_101 | tf_resnetv1_101_imagenet_224_224_14.4G | 23.55 | 42.4 | 91.5 |
| 49 | resnet_v1_152 | tf_resnetv1_152_imagenet_224_224_21.83G | 34.20 | 29.2 | 63.8 |
| 50 | vgg_16 | tf_vgg16_imagenet_224_224_30.96G | 50.17 | 19.9 | 40.9 |
| 51 | vgg_19 | tf_vgg19_imagenet_224_224_39.28G | 58.13 | 17.2 | 36.5 |
| 36 | plate detection | cf_plate-detection_320_320_0.49G | 2.00 | 500.0 | 1792.2 |
| 37 | plate recognition | cf_plate-recognition_96_288_1.75G | 8.82 | 113.4 | 383.2 |
| 38 | Inception_resnet_v2 | tf_inceptionresnetv2_imagenet_299_299_26.35G | 43.19 | 23.1 | 48.7 |
| 39 | Inception_v1 | tf_inceptionv1_imagenet_224_224_3G | 5.43 | 184.1 | 423.9 |
| 40 | Inception_v3 | tf_inceptionv3_imagenet_299_299_11.45G | 17.44 | 57.3 | 126.7 |
| 41 | Inception_v4 | tf_inceptionv4_imagenet_299_299_24.55G | 35.08 | 28.5 | 66.2 |
| 42 | Mobilenet_v1 | tf_mobilenetv1_0.25_imagenet_128_128_27.15M | 0.85 | 1170.7 | 4043.5 |
| 43 | Mobilenet_v1 | tf_mobilenetv1_0.5_imagenet_160_160_150.07M | 1.41 | 707.6 | 2007.1 |
| 44 | Mobilenet_v1 | tf_mobilenetv1_1.0_imagenet_224_224_1.14G | 3.52 | 284.3 | 754.9 |
| 45 | Mobilenet_v2 | tf_mobilenetv2_1.0_imagenet_224_224_0.59G | 4.33 | 230.8 | 568.4 |
| 46 | Mobilenet_v2 | tf_mobilenetv2_1.4_imagenet_224_224_1.16G | 5.98 | 167.3 | 393.1 |
| 47 | resnet_v1_50 | tf_resnetv1_50_imagenet_224_224_6.97G | 12.63 | 79.1 | 161.9 |
| 48 | resnet_v1_101 | tf_resnetv1_101_imagenet_224_224_14.4G | 23.16 | 43.1 | 91.3 |
| 49 | resnet_v1_152 | tf_resnetv1_152_imagenet_224_224_21.83G | 33.82 | 29.6 | 63.7 |
| 50 | vgg_16 | tf_vgg16_imagenet_224_224_30.96G | 49.78 | 20.1 | 40.9 |
| 51 | vgg_19 | tf_vgg19_imagenet_224_224_39.28G | 57.72 | 17.3 | 36.5 |
| 52 | ssd_mobilenet_v1 | tf_ssdmobilenetv1_coco_300_300_2.47G | 11.09 | 90.1 | 332.9 |
| 53 | ssd_mobilenet_v2 | tf_ssdmobilenetv2_coco_300_300_3.75G | 15.64 | 63.9 | 193.2 |
| 54 | ssd_resnet_50_v1_fpn | tf_ssdresnet50v1_fpn_coco_640_640_178.4G | 757.63 | 1.3 | 5.1 |
| 54 | ssd_resnet_50_v1_fpn | tf_ssdresnet50v1_fpn_coco_640_640_178.4G | 757.63 | 1.3 | 5.1 |
| 55 | yolov3_voc | tf_yolov3_voc_416_416_65.63G | 74.22 | 13.5 | 35.0 |
| 56 | mlperf_ssd_resnet34 | tf_mlperf_resnet34_coco_1200_1200_433G | 502.41 | 2.0 | 7.2 |
| 56 | mlperf_ssd_resnet34 | tf_mlperf_resnet34_coco_1200_1200_433G | 502.41 | 2.0 | 7.2 |

</details>

Expand Down

0 comments on commit 6a699e3

Please sign in to comment.