Faceswap feature causes the app to crash on Google Colab

[ruipedrodias94]

Describe the problem
When trying to use Faceswap feature in Google Colab, the app crashes.

Full Console Log
Image generated with private log at: /content/Fooocus/outputs/2023-12-04/log.html Generating and saving time: 30.31 seconds Requested to load SDXLClipModel Requested to load GPT2LMHeadModel Loading 2 new models [Fooocus Model Management] Moving model(s) has taken 0.85 seconds Total time: 70.66 seconds [Parameters] Adaptive CFG = 7 [Parameters] Sharpness = 2 [Parameters] ADM Scale = 1.5 : 0.8 : 0.3 [Parameters] CFG = 4.0 [Parameters] Seed = 51281143977912873 [Fooocus] Downloading control models ... [Fooocus] Loading control models ... ^C

[cad1231]

If you are using the free version of Google Colab, there is not enough RAM for faceswap. You need the paid version, and then go to 'change runtime type' and make sure that 'high RAM' is turned on. You don't need to change anything else. 50 compute units that you get with a monthly subscription should last you 20-30 hours of processing time.

[RukshanJS]

If you are using the free version of Google Colab, there is not enough RAM for faceswap. You need the paid version, and then go to 'change runtime type' and make sure that 'high RAM' is turned on. You don't need to change anything else. 50 compute units that you get with a monthly subscription should last you 20-30 hours of processing time.

That's a nice answer. Could someone kindly add this to the Readme under the colab section?

[tbuyle]

Try changing the last line of the colab with :

!python entry_with_update.py --share --preset --always-high-vram --all-in-fp16

See also #1377

[Damarcreative]

Always monitor RAM and GPU usage,
By default, using normal VRAM will make your RAM full and will stop.
since GPU usage is quite low, try adding --highvram parameter this will decrease RAM usage and increase GPU slightly.