-
Notifications
You must be signed in to change notification settings - Fork 13
/
get_flops.py
75 lines (62 loc) · 2.07 KB
/
get_flops.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
"""
An example to use fvcore to count MACs.
please install the following packages
`pip install timm fvcore`
Example command:
python get_flops.py --model moganet_tiny
"""
import argparse
import torch
from timm.models import create_model
from fvcore.nn import FlopCountAnalysis, flop_count_table
import models # register_model for MogaNet
def get_args():
parser = argparse.ArgumentParser()
parser.add_argument(
'--model',
type=str,
default='moganet_tiny',
help='model name')
parser.add_argument(
'--img_size',
type=int, default=224,
metavar='N',
help='Image patch size (default: None => model default)')
parser.add_argument(
'--throughput',
action='store_true', default=False,
help='Caculate throughput of model (default: False)')
args = parser.parse_args()
return args
def get_throughput(input, model):
_ = model(input[:2, ...]) # dummy forward
bs = 100
repetitions = 100
_, C, H, W = input.shape
input = torch.rand(bs, C, H, W).to("cuda")
model = model.to("cuda")
total_time = 0
with torch.no_grad():
for i in range(repetitions):
starter, ender = torch.cuda.Event(enable_timing=True), torch.cuda.Event(enable_timing=True)
starter.record()
_ = model(input)
ender.record()
torch.cuda.synchronize()
curr_time = starter.elapsed_time(ender) / 1000
total_time += curr_time
Throughput = (repetitions * bs) / total_time
print('Throughputs of {}: {:.3f}'.format(args.model, Throughput))
if __name__ == '__main__':
args = get_args()
model = create_model(args.model)
model.eval()
# print(model)
input = torch.rand(1, 3, args.img_size, args.img_size)
# Please note that FLOP here actually means MAC.
flop = FlopCountAnalysis(model, input)
print(flop_count_table(flop, max_depth=4))
print('MACs (G) of {}: {:.3f}'.format(args.model, flop.total() / 1e9))
# Get throughput
if args.throughput:
get_throughput(input, model)