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
|
from torchvision import models
import torch
from torch.backends._coreml.preprocess import CompileSpec, CoreMLComputeUnit, TensorSpec
def mobilenetv2_spec():
return {
"forward": CompileSpec(
inputs=(TensorSpec(shape=[1, 3, 224, 224]),),
outputs=(TensorSpec(shape=[1, 1000]),),
backend=CoreMLComputeUnit.CPU,
allow_low_precision=True,
),
}
def main():
model = models.mobilenet_v2(weights=models.MobileNet_V2_Weights.IMAGENET1K_V1)
model.eval()
example = torch.rand(1, 3, 224, 224)
model = torch.jit.trace(model, example)
compile_spec = mobilenetv2_spec()
mlmodel = torch._C._jit_to_backend("coreml", model, compile_spec)
print(mlmodel._c._get_method("forward").graph)
mlmodel._save_for_lite_interpreter("../models/model_coreml.ptl")
torch.jit.save(mlmodel, "../models/model_coreml.pt")
if __name__ == "__main__":
main()
|
