Closed shehrum closed 3 years ago
in the train script edit the code to this
try:
iteration = checkpoint_dict['iteration']
model_dict = checkpoint_dict['model'].state_dict()
except:
iteration = 1
model_dict = checkpoint_dict['state_dict']
Thankyou! Changed to this already and it worked.
I am trying to train on a custom dataset following the steps for
with the command:
python train.py -c config.json -p train_config.ignore_layers=["speaker_embedding.weight"] train_config.checkpoint_path="models/flowtron_ljs.pt"
However, I get the error:
It seems as the model doesn't have these keys. How should I go about fixing/modifying the training script to cater for that. Since the current
train.py
script throws this error. Would appreciate if anyone could help fix the issue.