You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Aug 30, 2018. It is now read-only.
I'm seeing the following error when loading official onnx models:
Traceback (most recent call last):
File "test_model_large_stepping.py", line 37, in test
cf_rep = c2.prepare(_model)
File "/Users/xxx/onnx-caffe2/onnx_caffe2/backend.py", line 559, in prepare
for imp in model.opset_import:
AttributeError: 'ModelProto' object has no attribute 'opset_import'
I'm using latest onnx from source, latest onnx-caffe2 from pip install.
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from __future__ import unicode_literals
import unittest
import numpy as np
import onnx
import onnx_caffe2.backend as c2
from onnx import helper
from onnx.onnx_pb2 import TensorProto
class TestLargeModel(unittest.TestCase):
MODEL_PATH = "../../../onnx_models/"
def test(self):
_model = onnx.load(self.MODEL_PATH + "shufflenet/model.pb")
cf_rep = c2.prepare(_model)
if __name__ == '__main__':
unittest.main()