I can't proceed to the second training

[ryugonomura]

I executed training as follows:

python train.py --train_data data_lmdb_release/training --valid_data data_lmdb_release/validation --select_data MJ-ST --batch_ratio 0.5-0.5 --Transformation TPS --FeatureExtraction ResNet --SequenceModeling BiLSTM --Prediction Attn

The first time was executed as shown below, but the step cannot proceed to 2/300000. Please tell me the cause?

------------ Options ------------- exp_name: TPS-ResNet-BiLSTM-Attn-Seed1111 train_data: data_lmdb_release/training valid_data: data_lmdb_release/validation manualSeed: 1111 workers: 4 batch_size: 192 num_iter: 300000 valInterval: 2000 saved_model: FT: False adam: False lr: 1 beta1: 0.9 rho: 0.95 eps: 1e-08 grad_clip: 5 baiduCTC: False select_data: ['MJ', 'ST'] batch_ratio: ['0.5', '0.5'] total_data_usage_ratio: 1.0 batch_max_length: 25 imgH: 32 imgW: 100 rgb: False character: 0123456789abcdefghijklmnopqrstuvwxyz sensitive: False PAD: False data_filtering_off: False Transformation: TPS FeatureExtraction: ResNet SequenceModeling: BiLSTM Prediction: Attn num_fiducial: 20 input_channel: 1 output_channel: 512 hidden_size: 256 num_gpu: 1 num_class: 38

[1/300000] Train loss: 3.63514, Valid loss: 3.57144, Elapsed_time: 1.75337 Current_accuracy : 0.000, Current_norm_ED : 0.03 Best_accuracy : 0.000, Best_norm_ED : 0.03

Ground Truth | Prediction | Confidence Score & T/F

on | iinnnnnnnnnninnninnninnni | 0.0000 False kishtwar | innnnnnnnnnnnnnnnnnnnnnnn | 0.0000 False crocs | aannnnnnnnnnnnnnnnnnnnnnn | 0.0000 False down | innnnnnnnnnnnooioioiinnnn | 0.0000 False block | aaaaaaaaaaaaaaaaaaaaaaaaa | 0.0000 False

[jusim]

Console output is printed each 2000 iterations (when validating). You can either change that with the --valInterval option or you just have to wait because it takes a while depending on the setup you use.