AttributeError: 'LSTMForecast' object has no attribute 'hidden_cell'

[wjd]

I‘ve encountered this error:

Traceback (most recent call last):
File "/home/kats/kats_prediction.py", line 214, in
fcst_df = predict_lstm(ts)
File "/home/kats/kats_prediction.py", line 53, in predict_lstm
fcst = lstm.predict(steps=60)
File "/home/miniconda3/envs/kats/lib/python3.7/site-packages/kats/models/lstm.py", line 236, in predict
test_inputs.append(self.model(seq).item())
File "/home/miniconda3/envs/kats/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/miniconda3/envs/kats/lib/python3.7/site-packages/kats/models/lstm.py", line 95, in forward
lstm_out, self.hidden_cell = self.lstm(input_seq.view(len(input_seq), 1, -1), self.hidden_cell)
File "/home/miniconda3/envs/kats/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1178, in getattr
type(self).name, name))
AttributeError: 'LSTMForecast' object has no attribute 'hidden_cell'

My torch version is 1.10.0

Could you please help me with this?