Error While Using SAE to train!

[lenibee]

I have attached the picture, I get the error and then nothing happens. I trained using DE and it was just fine. Why is this a thing?
Image-
============== Model Summary ===============
==                                        ==
==        Model name: SAE                 ==
==                                        ==
== Current iteration: 0                   ==
==                                        ==
==------------ Model Options -------------==
==                                        ==
==       target_iter: 80000               ==
==       random_flip: True                ==
==        resolution: 128                 ==
==         face_type: f                   ==
==        learn_mask: True                ==
==    optimizer_mode: 1                   ==
==             archi: df                  ==
==           ae_dims: 512                 ==
==         e_ch_dims: 42                  ==
==         d_ch_dims: 21                  ==
==        ca_weights: False               ==
==        pixel_loss: False               ==
==  face_style_power: 10.0                ==
==    bg_style_power: 10.0                ==
==           ct_mode: rct                 ==
==          clipgrad: False               ==
==        batch_size: 4                   ==
==                                        ==
==-------------- Running On --------------==
==                                        ==
==      Device index: 0                   ==
==              Name: GeForce GTX 1660 Ti ==
==              VRAM: 6.00GB              ==
==                                        ==
============================================

 

[Fappenberg]

OOM means out of memory...you have several options
1. Disable face style / by style power
2. Change optimizer mode to 2 to include your ram memory
3. Lower batch size
4. Disable learn mask (You don't need this unless your convert with learned, even then you would only need to enable this for a couple of iterations, also disabling this makes training run faster)

 

[lenibee]

OOM means out of memory...you have several options
1. Disable face style / by style power
2. Change optimizer mode to 2 to include your ram memory
3. Lower batch size
4. Disable learn mask (You don't need this unless your convert with learned, even then you would only need to enable this for a couple of iterations, also disabling this makes training run faster)

The 2nd approach is the best for me. Thanks! Also, dope profile picture