search

chervonij / DFL-Colab

DeepFaceLab fork which provides IPython Notebook to use DFL with Google Colab
1.05k stars 265 forks source link

Unable to Change Training Options in DFL-Colab #63

Open AMINEMM13 opened 10 months ago

AMINEMM13 commented 10 months ago

Hello,

I've been using DFL-Colab for deep learning tasks and I've encountered an issue where I'm unable to change the training options.

When I first started using DFL-Colab, I was able to set the training options before starting the training process. However, now when I attempt to start training, it skips the option setting stage and starts training immediately with the previous options.

Here's the model summary that is displayed:

============ Model Summary ============
==                                   ==
==            Model name: test_SAEHD ==
==                                   ==
==     Current iteration: 500        ==
==                                   ==
==---------- Model Options ----------==
==                                   ==
==            resolution: 128        ==
==             face_type: head       ==
==     models_opt_on_gpu: True       ==
==                 archi: liae-ud    ==
==               ae_dims: 256        ==
==                e_dims: 64         ==
==                d_dims: 64         ==
==           d_mask_dims: 22         ==
==       masked_training: True       ==
==       eyes_mouth_prio: True       ==
==           uniform_yaw: False      ==
==         blur_out_mask: False      ==
==             adabelief: False      ==
==            lr_dropout: n          ==
==           random_warp: False      ==
==      random_hsv_power: 0.0        ==
==       true_face_power: 0.0        ==
==      face_style_power: 0.0        ==
==        bg_style_power: 0.0        ==
==               ct_mode: none       ==
==              clipgrad: False      ==
==              pretrain: False      ==
==       autobackup_hour: 2          ==
== write_preview_history: True       ==
==           target_iter: 500        ==
==       random_src_flip: False      ==
==       random_dst_flip: False      ==
==            batch_size: 8          ==
==             gan_power: 2.5        ==
==        gan_patch_size: 16         ==
==              gan_dims: 16         ==
==                                   ==
==----------- Running On ------------==
==                                   ==
==          Device index: 0          ==
==                  Name: Tesla T4   ==
==                  VRAM: 13.34GB    ==
==                                   ==
=======================================

I've tried looking for a model.json file to delete as some sources suggested, but I couldn't find it in the workspace/model directory.

I would like to keep the previous training results but change the options for future training. How can I achieve this?