Got parameter has not been initialized error during fine tuning

I tried to Fine-tuning an ONNX model with MXNet/Gluon. I tried to fine-tune Inception_v2 model by following this link

Unfortunately, I got the following error:

~/anaconda3/lib/python3.7/site-packages/mxnet/gluon/ in list_ctx(self)
    537             if self._deferred_init:
    538                 return self._deferred_init[1]
--> 539             raise RuntimeError("Parameter '%s' has not been initialized"
    540         return self._ctx_list

RuntimeError: **Parameter 'conv1/7x7_s2/bn_mean_0' has not been initialized**

Any help would be appreciated!

@mgsmsobuj I think that’s a bug indeed, I managed to reproduce the problem. From what I can see in the onnx model params there are no running mean and var parameters provided. Not sure why or if it’s on purpose. Can you please create a github issue on the repo?