Faceswap Forum

Yes. You need to use tensorflow's summary_iterator to parse it. Example here: https://github.com/deepfakes/faceswap/blob/cbba53ea67cb48d8886041ae594ca2dfda48a4c8/lib/gui/stats.py#L68 You can also load the logs straight up into Tensorboard. You can google around how to do that. It's fairly straightfo...

I will push a fix at some point today which fixes this

So... I was wandering if the graph of the raw losses can indicate -in some way- how good the training set is: For examples... If I have a lot of spikes or -to say it in another way- a big oscillation around the smoothed loss... is this a sign that the trainer has issues learning and need more quali...

It is normal if the 1070 was trained on Tensorflow and the 2070 on plaidML. Ultimately, your install is a mess, and trouble shooting all the conflicts that have been introduced is going to be next to impossible. I suggest scrubbing clean (following ALL of the steps in the link) and re-installing: ht...

Huh? I'll need to look into that....

In the meantime, I assume you opened a .fsw file? If so, then you can manually open that file and delete the masker option, re-open the file in faceswap and you should then be able to select a masker

Locked due to split topic:

viewtopic.php?f=6&t=605

Unfortunately the console output doesn't tell us anything useful.

I may be misunderstanding, but this is built in.... We store logs in the model folder which can be loaded up in tensorboard.

You can parse these files to extract the data

The Nvidia library is failing and it's falling back to plaidml. You shouldn't even have plaidml installed, so remove that.

Once done, It'll probably fail, but at least you'll get a reason why.

Also, post output of help > output system info please.

Help > Update Faceswap...

Restart Faceswap

It shouldn't be looking for plaidml then.

I suggest you delete your faceswap folder and re-run the installer, making sure you select the "NVIDIA" option when prompted.

You don't have plaidml installed. How did you install Faceswap. What OS? What GPU?

Ok, so you have a video you want to swap faces for. This video should have an associated (cleaned) alignments file. If you don't have one, go and create one now (following the standard Extract workflow). The path to this video is entered as input The path to the model folder is entered as model_dir ...

We don't have a guide.

What are you specifically stuck on. Convert should be relatively straight forward.

app.php/faqpage#f3r3

This happens to me occasionally in ubuntu too. I think it is an internal issue somewhere.

I find that if I close the preview tool and re-open it, previews start working :/

Unfortunately we don't have availability of a Mac to test against (and certainly don't have an OSX/Nvidia combo to test against), but we have never seen this error reported. This looks like an Apple specific error due to a badly compiled program, which may be because you are on a slightly older macO...

At this point, I'm not sure. Sorry. Ultimately in your screen grab above faceswap.py IS found as it outputs the help message (it needs to execute faceswap to do this). If you are getting a message that the file is not found, then you are most likely in the wrong folder. The main issue, I don't know....

2 and 3: app.php/faqpage#f0r3

For 1, your computer should shut itself down before it catches fire. If your machine is well ventilated and well powered, you should be fine.

1) Can I flip individual frames within the folder and tweak alignments without re-extracting everything from scratch? For example, can I use Update-Hashes after flipping? Or, probably more accurately, can I remove and purge the images to be flipped from the large Initial training set to create a se...

Faceswap Forum

Search found 2668 matches

Re: recording and graphiing training history

Re: Cannot Extract any more:

Re: Quality of the training set.

Re: Is this normal operation?

Re: Cannot Extract any more:

Re: ValueError: Unable to import backend : plaidml.keras.backend - How Can I solve it?

Re: Error 3221225477... maybe Fixed.

Re: recording and graphiing training history

Re: ValueError: Unable to import backend : plaidml.keras.backend - How Can I solve it?

Re: ValueError: Unable to import backend : plaidml.keras.backend - How Can I solve it?

Re: ValueError: Unable to import backend : plaidml.keras.backend - How Can I solve it?

Re: ValueError: Unable to import backend : plaidml.keras.backend - How Can I solve it?

Re: Convert - but how?

Re: Convert - but how?

Re: Can I save the train model?

Re: Getting blank preview results in XUbuntu.

Re: Illegal instruction: 4

Re: "command not found"

Re: Question about using GPU

Re: Training for asymetric facial features