TypeError: Cannot interpret feed_dict key as Tensor: Tensor Tensor("Placeholder:0", shape=(3, 3, 3, 64), dtype=float32) is not an element of this graph.

[ManjeeraJagiri]

When I try running camera_test.py,
If I have restarted my kernel, it works perfectly fine. But If i haven't restarted, I get the following error--
TypeError: Cannot interpret feed_dict key as Tensor: Tensor Tensor("Placeholder:0", shape=(3, 3, 3, 64), dtype=float32) is not an element of this graph.

Any idea as to why this is happening. Does it have to do anything with multithreading??

[asharsid]

I am getting the same error. App after restart works perfectly fine. Someone help please.