Why is learning no longer going on?

[leeisack]

While learning, the learning does not proceed any further without errors. I want to know why.

My training command is:

python scripts/train.py --dataset_type ffhq_encode --exp_dir ./ffhq_result2/ --start_from_latent_avg --use_w_pool --w_discriminator_lambda 0.1 --progressive_start 20000 --id_lambda 0.5 --val_interval 10000 --max_steps 250000 --stylegan_size 1024 --stylegan_weights ./stylegan2-ffhq-config-f.pt --workers 8 --batch_size 8 --test_batch_size 4 --test_workers 4

My training parameters are:

{ "batch_size": 8, "board_interval": 50, "checkpoint_path": null, "d_reg_every": 16, "dataset_type": "ffhq_encode", "delta_norm": 2, "delta_norm_lambda": 0.0002, "encoder_type": "Encoder4Editing", "exp_dir": "./ffhq_result2/", "id_lambda": 0.5, "image_interval": 100, "keep_optimizer": false, "l2_lambda": 1.0, "learning_rate": 0.0001, "lpips_lambda": 0.8, "lpips_type": "alex", "max_steps": 250000, "optim_name": "ranger", "progressive_start": 20000, "progressive_step_every": 2000, "progressive_steps": [ 0, 20000, 22000, 24000, 26000, 28000, 30000, 32000, 34000, 36000, 38000, 40000, 42000, 44000, 46000, 48000, 50000, 52000 ], "r1": 10, "resume_training_from_ckpt": null, "save_interval": null, "save_training_data": false, "start_from_latent_avg": true, "stylegan_size": 1024, "stylegan_weights": "./stylegan2-ffhq-config-f.pt", "sub_exp_dir": null, "test_batch_size": 4, "test_workers": 4, "train_decoder": false, "update_param_list": null, "use_w_pool": true, "val_interval": 10000, "w_discriminator_lambda": 0.1, "w_discriminator_lr": 2e-05, "w_pool_size": 50, "workers": 8 }

[leeisack]

I'm sorry. Because the validation dataset is very large, I misunderstood that it was not possible to proceed further.