pretrained model for fine-tuning

[kimlia545]

I added more character('train.py' --character) and got an error (size mismatch)
I don't know why It works when the Prediction is different from what pretrained model's Prediction
for example these code is work

python train.py --train_data data_lmdb/training --workers 0 --valid_data data_lmdb/validation --Transformation TPS --FeatureExtraction ResNet --SequenceModeling BiLSTM --Prediction CTC --saved_model pretrained_models/TPS-ResNet-BiLSTM-Attn.pth --FT

python train.py --train_data data_lmdb/training --workers 0 --valid_data data_lmdb/validation --Transformation TPS --FeatureExtraction ResNet --SequenceModeling BiLSTM --Prediction Attn --saved_model pretrained_models/TPS-ResNet-BiLSTM-CTC.pth --FT

but If Prediction is the same as pretrained_model has

python train_test.py --train_data data_lmdb/training --workers 0 --valid_data data_lmdb/validation --Transformation TPS --FeatureExtraction ResNet --SequenceModeling BiLSTM --Prediction Attn --saved_model pretrained_models/TPS-ResNet-BiLSTM-Attn.pth --FT

RuntimeError: Error(s) in loading state_dict for DataParallel:
size mismatch for module.Prediction.attention_cell.rnn.weight_ih: copying a param with shape torch.Size([1024, 294]) from checkpoint, the shape in current model is torch.Size([1024, 1637]).
size mismatch for module.Prediction.generator.weight: copying a param with shape torch.Size([38, 256]) from checkpoint, the shape in current model is torch.Size([1381, 256]).
size mismatch for module.Prediction.generator.bias: copying a param with shape torch.Size([38]) from checkpoint, the shape in current model is torch.Size([1381]).

and How to fix this error
'demo.py'
confidence_score = pred_max_prob.cumprod(dim=0)[-1]
IndexError: index -1 is out of bounds for dimension 0 with size 0