Before run the below code, check the config file. For training a path generator
cd learning-generator
sh run_dokg_multi_di_rev_no_hiar.sh
If you set the setting 'split_dataset' as True, split dataset for training by ratio of 0.9:0.05:0.05 will be saved under output_dir.
여기에서 data_dir, 그리고 train.txt는 무엇을 넣어야 하는지요?
Before run the below code, check the config file. For training a path generator
cd learning-generator
sh run_dokg_multi_di_rev_no_hiar.sh
If you set the setting 'split_dataset' as True, split dataset for training by ratio of 0.9:0.05:0.05 will be saved under output_dir.
여기에서 data_dir, 그리고 train.txt는 무엇을 넣어야 하는지요?