I am trying to test a trained model on some new test data.
I load the model as below:
model = ModelClass(input_dim=input_dim, vocab_size=vocab_size, model_config=model_config)
Then I load the model_state_dict
into this newly created model.
model_state_dict = torch.load(model_path)['model_state_dict']
Everything works fine till here.
When I however run a minibatch of the test data through this model, I get a
RuntimeError: parameter types mismatch
. This error is originating from the forward pass of the Encoder RNN. Below is the Traceback:
File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 491, in __call__
result = self.forward(*input, **kwargs)
File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/rnn.py", line 192, in forward
output, hidden = func(input, self.all_weights, hx, batch_sizes)
File "/usr/local/lib/python3.6/dist-packages/torch/nn/_functions/rnn.py", line 323, in forward
return func(input, *fargs, **fkwargs)
File "/usr/local/lib/python3.6/dist-packages/torch/nn/_functions/rnn.py", line 287, in forward
dropout_ts)
The data generating process is the same. I do not see where a types mismatch
could come from. Can some one help me in debugging this?
Thanks!