Alnex.net test error

[dsp6414]

Describe the bug
RuntimeError: [ONNXRuntimeError] : 1 : GENERAL ERROR : No suitable kernel definition found for op Dropout(6)

System information

• windows 10:
• onnxruntime-0.1.4-cp36-cp36m-win_amd64.whl:
• ONNX Runtime version:0.1.4
• Python version:3.6

code:
from torch.autograd import Variable
import torch.onnx
import torchvision
import os
def get_runningfile_path():
return os.path.dirname(os.path.abspath(file))
root_path = get_runningfile_path()
dummy_input = Variable(torch.randn(1, 3, 224, 224))
model = torchvision.models.alexnet(pretrained=True)
torch.onnx.export(model, dummy_input,os.path.join(root_path,'alexnet.onnx'))

Compute the prediction with ONNX Runtime

import onnxruntime as rt
import numpy

img = numpy.random.rand(1,3,224,224)
sess = rt.InferenceSession(os.path.join(root_path,'alexnet.onnx'))
input_name = sess.get_inputs()[0].name
label_name = sess.get_outputs()[0].name
pred_onx = sess.run([label_name], {input_name: img.astype(numpy.float32)})[0]
print(pred_onx)

[jywu-msft]

onnxruntime only supports opset 7 and later.
the error you are encountering is due to the model containing opset 6 ops, specifically Dropout-6

[snnn]

Close it because we have no plan to support opset 6.