遇到错误

[LAUGH111]

请问一下这个错误如何解决

[lipengfeizju]

TL;DR: You need to input the required arguments for run.py when you run the program.

Please refer to the link here for more deatils

https://github.com/Ren-Research/LOMAR/blob/main/TUTORIAL.md

If you want to use the code, you should follow these steps:

1.Generate graph dataset 2.Train the RL model 3.Evaluate the policy

More specifically, when you use python run.py, at least you need to tell the program wether you want to train the model or evaluate the model. A quick example for training is shown as below

python run.py --encoder mpnn --model inv-ff-hist --problem adwords --batch_size 100 --embedding_dim 30 --n_heads 1 --u_size 10  --v_size 60 \
            --n_epochs 20 --train_dataset dataset/train/adwords_triangular_uniform_0.10.4_10by60/parameter_-1 \
            --val_dataset dataset/val/adwords_triangular_uniform_0.10.4_10by60/parameter_-1 \
            --dataset_size 1000 --val_size 100 --checkpoint_epochs 0 --baseline exponential --lr_model 0.006 --lr_decay 0.97 \
            --output_dir saved_models --log_dir logs_dataset --n_encode_layers 1 \
            --save_dir saved_models/adwords_triangular_uniform_0.10.4_10by60/parameter_-1 \
            --graph_family_parameter -1 --exp_beta 0.8 --ent_rate 0.0006

But only after you finish the data generation, this should can be sucessfully executed.

So please refer to the TUTORIAL.md file for more details.