Torch.jitted.script_method, and load_state_dict

I have a function annotated with torch.jit.script_method whose forward pass includes for dropout

class Convs(torch.jit.ScriptModule):
    def __init__(self):
        self.conv = nn.Conv(etc)
    def forward(self, x, input_lengths):
        for conv in self.convolutions:
            x = F.dropout(F.relu(self.conv(x)), 0.5,

When loading a state_dict onto the jitted model, there’s a Missing Key error

Traceback (most recent call last):
  File "", line 54, in <module>
  File "/opt/conda/lib/python3.6/site-packages/torch/nn/modules/", line 769, in load_state_dict
    self.__class__.__name__, "\n\t".join(error_msgs)))
RuntimeError: Error(s) in loading state_dict for Tacotron2:
	Missing key(s) in state_dict: "".

How should be used with torch.jit such that the behavior mimics non-jit pytorch?