Hi Everybody,
First of all, awesome project/library
Now my question…
If I am not mistaken I trained my model with fastai-1.0.50, now I am would like to use my trained model with version fastai-1.0.53, but it has model discrepancy.
here is my learner:
awd_lstm_lm_config['qrnn'] = True
learn = language_model_learner(tickets_lm, arch=AWD_LSTM,
pretrained=True,
drop_mult=0.3,
callback_fns=ShowGraph,
pretrained_fnames=['model-60vocab','full_lm_60k_itos'])
I am getting this error:
RuntimeError: Error(s) in loading state_dict for SequentialRNN:
Missing key(s) in state_dict: "0.rnns.0.layers.0.linear.weight_raw", "0.rnns.0.layers.0.linear.module.weight", "0.rnns.0.layers.0.linear.module.bias", "0.rnns.1.layers.0.linear.weight_raw", "0.rnns.1.layers.0.linear.module.weight", "0.rnns.1.layers.0.linear.module.bias", "0.rnns.2.layers.0.linear.weight_raw", "0.rnns.2.layers.0.linear.module.weight", "0.rnns.2.layers.0.linear.module.bias".
Unexpected key(s) in state_dict: "0.rnns.0.weight_hh_l0_raw", "0.rnns.0.module.weight_ih_l0", "0.rnns.0.module.weight_hh_l0", "0.rnns.0.module.bias_ih_l0", "0.rnns.0.module.bias_hh_l0", "0.rnns.1.weight_hh_l0_raw", "0.rnns.1.module.weight_ih_l0", "0.rnns.1.module.weight_hh_l0", "0.rnns.1.module.bias_ih_l0", "0.rnns.1.module.bias_hh_l0", "0.rnns.2.weight_hh_l0_raw", "0.rnns.2.module.weight_ih_l0", "0.rnns.2.module.weight_hh_l0", "0.rnns.2.module.bias_ih_l0", "0.rnns.2.module.bias_hh_l0".
So, is it possible to convert my old model? or do I have to configure my new learner? or do I need to train from scratch?
Thank you
Br,
Marek