Skip to content

Commit 81ace22

Browse files
committed
Update benchmark in readme
1 parent 606d30d commit 81ace22

File tree

1 file changed

+22
-22
lines changed

1 file changed

+22
-22
lines changed

README.md

Lines changed: 22 additions & 22 deletions
Original file line numberDiff line numberDiff line change
@@ -96,28 +96,28 @@ If ptflops was useful for your paper or tech report, please cite me:
9696

9797
### [torchvision](https://pytorch.org/vision/0.16/models.html)
9898

99-
Model | Input Resolution | Params(M) | MACs(G)
100-
--- |--- |--- |---
101-
alexnet | 224x224 | 61.10 | 0.72
102-
convnext_base | 224x224 | 88.59 | 15.43
103-
densenet121 | 224x224 | 7.98 | 2.90
104-
efficientnet_b0 | 224x224 | 5.29 | 0.41
105-
efficientnet_v2_m | 224x224 | 54.14 | 5.43
106-
googlenet | 224x224 | 13.00 | 1.51
107-
inception_v3 | 224x224 | 27.16 | 2.86
108-
maxvit_t | 224x224 | 30.92 | 5.48
109-
mnasnet1_0 | 224x224 | 4.38 | 0.33
110-
mobilenet_v2 | 224x224 | 3.50 | 0.32
111-
mobilenet_v3_large | 224x224 | 5.48 | 0.23
112-
regnet_y_1_6gf | 224x224 | 11.20 | 1.65
113-
resnet18 | 224x224 | 11.69 | 1.83
114-
resnet50 | 224x224 | 25.56 | 4.13
115-
resnext50_32x4d | 224x224 | 25.03 | 4.29
116-
shufflenet_v2_x1_0 | 224x224 | 2.28 | 0.15
117-
squeezenet1_0 | 224x224 | 1.25 | 0.84
118-
vgg16 | 224x224 | 138.36 | 15.52
119-
vit_b_16 | 224x224 | 86.57 | 17.60
120-
wide_resnet50_2 | 224x224 | 68.88 | 11.45
99+
Model | Input Resolution | Params(M) | MACs(G) (`pytorch`) | MACs(G) (`aten`)
100+
--- |--- |--- |--- |---
101+
alexnet | 224x224 | 61.10 | 0.72 | 0.71
102+
convnext_base | 224x224 | 88.59 | 15.43 | 15.38
103+
densenet121 | 224x224 | 7.98 | 2.90 |
104+
efficientnet_b0 | 224x224 | 5.29 | 0.41 |
105+
efficientnet_v2_m | 224x224 | 54.14 | 5.43 |
106+
googlenet | 224x224 | 13.00 | 1.51 |
107+
inception_v3 | 224x224 | 27.16 | 5.75 | 5.71
108+
maxvit_t | 224x224 | 30.92 | 5.48 |
109+
mnasnet1_0 | 224x224 | 4.38 | 0.33 |
110+
mobilenet_v2 | 224x224 | 3.50 | 0.32 |
111+
mobilenet_v3_large | 224x224 | 5.48 | 0.23 |
112+
regnet_y_1_6gf | 224x224 | 11.20 | 1.65 |
113+
resnet18 | 224x224 | 11.69 | 1.83 | 1.81
114+
resnet50 | 224x224 | 25.56 | 4.13 | 4.09
115+
resnext50_32x4d | 224x224 | 25.03 | 4.29 |
116+
shufflenet_v2_x1_0 | 224x224 | 2.28 | 0.15 |
117+
squeezenet1_0 | 224x224 | 1.25 | 0.84 | 0.82
118+
vgg16 | 224x224 | 138.36 | 15.52 | 15.48
119+
vit_b_16 | 224x224 | 86.57 | 17.61 (wrong) | 16.86
120+
wide_resnet50_2 | 224x224 | 68.88 | 11.45 |
121121

122122

123123
### [timm](https://github.com/huggingface/pytorch-image-models)

0 commit comments

Comments
 (0)