mirror of
https://github.com/deepfakes/faceswap.git
synced 2026-10-18 14:34:30 +08:00
torzdf
837bc2d51d
* FaceSwap 3 (#1515) * Update extract pipeline * Update requirements + setup for nvidia * Remove allow-growth option * tf.keras to keras updates * lib.model.losses - Port + fix all loss functions for Keras3 * lib.model - port initializers, layers. normalization to Keras3 * lib.model.autoclip to Keras 3 * Update mixed precision layer storage * model file to .keras format * Restructure nn_blocks to initialize layers in __init__ * Tensorboard - Trainer: Add Torch compatible Tensorboard callbacks - GUI event reader remove TF dependency * Loss logging - Flush TB logs on save - Replace TB live iterator for GUI * Backup models on total loss drop rather than per side * Update all models to Keras3 Compat * Remove lib.model.session * Update clip ViT to Keras 3 * plugins.extract.mask.unet-dfl - Fix for Keras3/Torch backend * Port AdaBelief to Keras 3 * setup.py: - Add --dev flag for dev tool install * Fix Keras 3 syntax * Fix LR Finder for Keras 3 * Fix mixed precision switching for Keras 3 * Add more optimizers + open up config setting * train: Remove updating FS1 weights to FS2 models * Alignments: Remove support for legacy .json files * tools.model: - Remove TF Saved Format saving - Fix Backup/Restore + Nan-Scan * Fix inference model creation for Keras 3 * Preview tool: Fix for Keras3 * setup.py: Configure keras backend * train: Migration of FS2 models to FS3 * Training: Default coverage to 100% * Remove DirectML backend * Update setup for MacOS * GUI: Force line reading to UTF-8 * Remove redundant Tensorflow references * Remove redundant code * Legacy model loading: Fix TFLamdaOp scalar ops and DepthwiseConv2D * Add vertical offset option for training * Github actions: Add more python versions * Add python version to workflow names * Github workflow: Exclude Python 3.12 for macOS * Implement custom training loop * Fs3 - Add RTX5xxx and ROCm 6.1-6.4 support (#1511) * setup.py: Add Cuda/ROCm version select options * bump minimum python version to 3.11 * Switch from setup.cgf to pyproject.toml * Documentation: Update all docs to use automodapi * Allow sysinfo to run with missing packages + correctly install tk under Linux * Bugfix: dot naming convention in clip models * lib.config: Centralise globally rather than passing as object - Add torch DataParallel for multi-gpu training - GUI: Group switches together when generating cli args - CLI: Remove deprecated multi-character argparse args - Refactor: - Centralise tensorboard reading/writing + unit tests - Create trainer plugin interfaces + add original + distributed * Update installers
76 lines
2.9 KiB
Python
76 lines
2.9 KiB
Python
#!/usr/bin/env python3
|
|
""" Improved autoencoder for faceswap """
|
|
|
|
from keras import Input, layers, Model as KModel
|
|
|
|
from lib.model.nn_blocks import Conv2DOutput, Conv2DBlock, UpscaleBlock
|
|
from plugins.train.train_config import Loss as cfg_loss
|
|
|
|
from ._base import ModelBase
|
|
# pylint:disable=duplicate-code
|
|
|
|
|
|
class Model(ModelBase):
|
|
""" Improved Autoencoder Model """
|
|
def __init__(self, *args, **kwargs):
|
|
super().__init__(*args, **kwargs)
|
|
self.input_shape = (64, 64, 3)
|
|
self.encoder_dim = 1024
|
|
|
|
def build_model(self, inputs):
|
|
""" Build the IAE Model """
|
|
encoder = self.encoder()
|
|
decoder = self.decoder()
|
|
inter_a = self.intermediate("a")
|
|
inter_b = self.intermediate("b")
|
|
inter_both = self.intermediate("both")
|
|
|
|
encoder_a = encoder(inputs[0])
|
|
encoder_b = encoder(inputs[1])
|
|
|
|
outputs = (decoder(layers.Concatenate()([inter_a(encoder_a), inter_both(encoder_a)])) +
|
|
decoder(layers.Concatenate()([inter_b(encoder_b), inter_both(encoder_b)])))
|
|
|
|
autoencoder = KModel(inputs, outputs, name=self.model_name)
|
|
return autoencoder
|
|
|
|
def encoder(self):
|
|
""" Encoder Network """
|
|
input_ = Input(shape=self.input_shape)
|
|
var_x = input_
|
|
var_x = Conv2DBlock(128, activation="leakyrelu")(var_x)
|
|
var_x = Conv2DBlock(256, activation="leakyrelu")(var_x)
|
|
var_x = Conv2DBlock(512, activation="leakyrelu")(var_x)
|
|
var_x = Conv2DBlock(1024, activation="leakyrelu")(var_x)
|
|
var_x = layers.Flatten()(var_x)
|
|
return KModel(input_, var_x, name="encoder")
|
|
|
|
def intermediate(self, side):
|
|
""" Intermediate Network """
|
|
input_ = Input(shape=(4 * 4 * 1024, ))
|
|
var_x = layers.Dense(self.encoder_dim)(input_)
|
|
var_x = layers.Dense(4 * 4 * int(self.encoder_dim/2))(var_x)
|
|
var_x = layers.Reshape((4, 4, int(self.encoder_dim/2)))(var_x)
|
|
return KModel(input_, var_x, name=f"inter_{side}")
|
|
|
|
def decoder(self):
|
|
""" Decoder Network """
|
|
input_ = Input(shape=(4, 4, self.encoder_dim))
|
|
var_x = input_
|
|
var_x = UpscaleBlock(512, activation="leakyrelu")(var_x)
|
|
var_x = UpscaleBlock(256, activation="leakyrelu")(var_x)
|
|
var_x = UpscaleBlock(128, activation="leakyrelu")(var_x)
|
|
var_x = UpscaleBlock(64, activation="leakyrelu")(var_x)
|
|
var_x = Conv2DOutput(3, 5, name="face_out")(var_x)
|
|
outputs = [var_x]
|
|
|
|
if cfg_loss.learn_mask():
|
|
var_y = input_
|
|
var_y = UpscaleBlock(512, activation="leakyrelu")(var_y)
|
|
var_y = UpscaleBlock(256, activation="leakyrelu")(var_y)
|
|
var_y = UpscaleBlock(128, activation="leakyrelu")(var_y)
|
|
var_y = UpscaleBlock(64, activation="leakyrelu")(var_y)
|
|
var_y = Conv2DOutput(1, 5, name="mask_out")(var_y)
|
|
outputs.append(var_y)
|
|
return KModel(input_, outputs=outputs, name="decoder")
|