1
0
Fork 0
mirror of https://github.com/deepfakes/faceswap synced 2025-06-07 10:37:19 -04:00
faceswap/lib/gui/display_command.py
torzdf cd00859c40
model_refactor (#571) (#572)
* model_refactor (#571)

* original model to new structure

* IAE model to new structure

* OriginalHiRes to new structure

* Fix trainer for different resolutions

* Initial config implementation

* Configparse library added

* improved training data loader

* dfaker model working

* Add logging to training functions

* Non blocking input for cli training

* Add error handling to threads. Add non-mp queues to queue_handler

* Improved Model Building and NNMeta

* refactor lib/models

* training refactor. DFL H128 model Implementation

* Dfaker - use hashes

* Move timelapse. Remove perceptual loss arg

* Update INSTALL.md. Add logger formatting. Update Dfaker training

* DFL h128 partially ported

* Add mask to dfaker (#573)

* Remove old models. Add mask to dfaker

* dfl mask. Make masks selectable in config (#575)

* DFL H128 Mask. Mask type selectable in config.

* remove gan_v2_2

* Creating Input Size config for models

Creating Input Size config for models

Will be used downstream in converters.

Also name change of image_shape to input_shape to clarify ( for future models with potentially different output_shapes)

* Add mask loss options to config

* MTCNN options to config.ini. Remove GAN config. Update USAGE.md

* Add sliders for numerical values in GUI

* Add config plugins menu to gui. Validate config

* Only backup model if loss has dropped. Get training working again

* bugfixes

* Standardise loss printing

* GUI idle cpu fixes. Graph loss fix.

* mutli-gpu logging bugfix

* Merge branch 'staging' into train_refactor

* backup state file

* Crash protection: Only backup if both total losses have dropped

* Port OriginalHiRes_RC4 to train_refactor (OriginalHiRes)

* Load and save model structure with weights

* Slight code update

* Improve config loader. Add subpixel opt to all models. Config to state

* Show samples... wrong input

* Remove AE topology. Add input/output shapes to State

* Port original_villain (birb/VillainGuy) model to faceswap

* Add plugin info to GUI config pages

* Load input shape from state. IAE Config options.

* Fix transform_kwargs.
Coverage to ratio.
Bugfix mask detection

* Suppress keras userwarnings.
Automate zoom.
Coverage_ratio to model def.

* Consolidation of converters & refactor (#574)

* Consolidation of converters & refactor

Initial Upload of alpha

Items
- consolidate convert_mased & convert_adjust into one converter
-add average color adjust to convert_masked
-allow mask transition blur size to be a fixed integer of pixels and a fraction of the facial mask size
-allow erosion/dilation size to be a fixed integer of pixels and a fraction of the facial mask size
-eliminate redundant type conversions to avoid multiple round-off errors
-refactor loops for vectorization/speed
-reorganize for clarity & style changes

TODO
- bug/issues with warping the new face onto a transparent old image...use a cleanup mask for now
- issues with mask border giving black ring at zero erosion .. investigate
- remove GAN ??
- test enlargment factors of umeyama standard face .. match to coverage factor
- make enlargment factor a model parameter
- remove convert_adjusted and referencing code when finished

* Update Convert_Masked.py

default blur size of 2 to match original...
description of enlargement tests
breakout matrxi scaling into def

* Enlargment scale as a cli parameter

* Update cli.py

* dynamic interpolation algorithm

Compute x & y scale factors from the affine matrix on the fly by QR decomp.
Choose interpolation alogrithm for the affine warp based on an upsample or downsample for each image

* input size
input size from config

* fix issues with <1.0 erosion

* Update convert.py

* Update Convert_Adjust.py

more work on the way to merginf

* Clean up help note on sharpen

* cleanup seamless

* Delete Convert_Adjust.py

* Update umeyama.py

* Update training_data.py

* swapping

* segmentation stub

* changes to convert.str

* Update masked.py

* Backwards compatibility fix for models
Get converter running

* Convert:
Move masks to class.
bugfix blur_size
some linting

* mask fix

* convert fixes

- missing facehull_rect re-added
- coverage to %
- corrected coverage logic
- cleanup of gui option ordering

* Update cli.py

* default for blur

* Update masked.py

* added preliminary low_mem version of OriginalHighRes model plugin

* Code cleanup, minor fixes

* Update masked.py

* Update masked.py

* Add dfl mask to convert

* histogram fix & seamless location

* update

* revert

* bugfix: Load actual configuration in gui

* Standardize nn_blocks

* Update cli.py

* Minor code amends

* Fix Original HiRes model

* Add masks to preview output for mask trainers
refactor trainer.__base.py

* Masked trainers converter support

* convert bugfix

* Bugfix: Converter for masked (dfl/dfaker) trainers

* Additional Losses (#592)

* initial upload

* Delete blur.py

* default initializer = He instead of Glorot (#588)

* Allow kernel_initializer to be overridable

* Add ICNR Initializer option for upscale on all models.

* Hopefully fixes RSoDs with original-highres model plugin

* remove debug line

* Original-HighRes model plugin Red Screen of Death fix, take #2

* Move global options to _base. Rename Villain model

* clipnorm and res block biases

* scale the end of res block

* res block

* dfaker pre-activation res

* OHRES pre-activation

* villain pre-activation

* tabs/space in nn_blocks

* fix for histogram with mask all set to zero

* fix to prevent two networks with same name

* GUI: Wider tooltips. Improve TQDM capture

* Fix regex bug

* Convert padding=48 to ratio of image size

* Add size option to alignments tool extract

* Pass through training image size to convert from model

* Convert: Pull training coverage from model

* convert: coverage, blur and erode to percent

* simplify matrix scaling

* ordering of sliders in train

* Add matrix scaling to utils. Use interpolation in lib.aligner transform

* masked.py Import get_matrix_scaling from utils

* fix circular import

* Update masked.py

* quick fix for matrix scaling

* testing thus for now

* tqdm regex capture bugfix

* Minor ammends

* blur size cleanup

* Remove coverage option from convert (Now cascades from model)

* Implement convert for all model types

* Add mask option and coverage option to all existing models

* bugfix for model loading on convert

* debug print removal

* Bugfix for masks in dfl_h128 and iae

* Update preview display. Add preview scaling to cli

* mask notes

* Delete training_data_v2.py

errant file

* training data variables

* Fix timelapse function

* Add new config items to state file for legacy purposes

* Slight GUI tweak

* Raise exception if problem with loaded model

* Add Tensorboard support (Logs stored in model directory)

* ICNR fix

* loss bugfix

* convert bugfix

* Move ini files to config folder. Make TensorBoard optional

* Fix training data for unbalanced inputs/outputs

* Fix config "none" test

* Keep helptext in .ini files when saving config from GUI

* Remove frame_dims from alignments

* Add no-flip and warp-to-landmarks cli options

* Revert OHR to RC4_fix version

* Fix lowmem mode on OHR model

* padding to variable

* Save models in parallel threads

* Speed-up of res_block stability

* Automated Reflection Padding

* Reflect Padding as a training option

Includes auto-calculation of proper padding shapes, input_shapes, output_shapes

Flag included in config now

* rest of reflect padding

* Move TB logging to cli. Session info to state file

* Add session iterations to state file

* Add recent files to menu. GUI code tidy up

* [GUI] Fix recent file list update issue

* Add correct loss names to TensorBoard logs

* Update live graph to use TensorBoard and remove animation

* Fix analysis tab. GUI optimizations

* Analysis Graph popup to Tensorboard Logs

* [GUI] Bug fix for graphing for models with hypens in name

* [GUI] Correctly split loss to tabs during training

* [GUI] Add loss type selection to analysis graph

* Fix store command name in recent files. Switch to correct tab on open

* [GUI] Disable training graph when 'no-logs' is selected

* Fix graphing race condition

* rename original_hires model to unbalanced
2019-02-09 18:35:12 +00:00

234 lines
9.1 KiB
Python

#!/usr/bin python3
""" Command specific tabs of Display Frame of the Faceswap GUI """
import datetime
import logging
import os
import tkinter as tk
from tkinter import ttk
from .display_graph import TrainingGraph
from .display_page import DisplayOptionalPage
from .tooltip import Tooltip
from .stats import Calculations
from .utils import FileHandler, get_config, get_images
logger = logging.getLogger(__name__) # pylint: disable=invalid-name
class PreviewExtract(DisplayOptionalPage): # pylint: disable=too-many-ancestors
""" Tab to display output preview images for extract and convert """
def display_item_set(self):
""" Load the latest preview if available """
logger.trace("Loading latest preview")
get_images().load_latest_preview()
self.display_item = get_images().previewoutput
def display_item_process(self):
""" Display the preview """
logger.trace("Displaying preview")
if not self.subnotebook.children:
self.add_child()
else:
self.update_child()
def add_child(self):
""" Add the preview label child """
logger.debug("Adding child")
preview = self.subnotebook_add_page(self.tabname, widget=None)
lblpreview = ttk.Label(preview, image=get_images().previewoutput[1])
lblpreview.pack(side=tk.TOP, anchor=tk.NW)
Tooltip(lblpreview, text=self.helptext, wraplength=200)
def update_child(self):
""" Update the preview image on the label """
logger.trace("Updating preview")
for widget in self.subnotebook_get_widgets():
widget.configure(image=get_images().previewoutput[1])
def save_items(self):
""" Open save dialogue and save preview """
location = FileHandler("dir", None).retfile
if not location:
return
filename = "extract_convert_preview"
now = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
filename = os.path.join(location,
"{}_{}.{}".format(filename,
now,
"png"))
get_images().previewoutput[0].save(filename)
logger.debug("Saved preview to %s", filename)
print("Saved preview to {}".format(filename))
class PreviewTrain(DisplayOptionalPage): # pylint: disable=too-many-ancestors
""" Training preview image(s) """
def __init__(self, *args, **kwargs):
self.update_preview = get_config().tk_vars["updatepreview"]
super().__init__(*args, **kwargs)
def display_item_set(self):
""" Load the latest preview if available """
logger.trace("Loading latest preview")
if not self.update_preview.get():
logger.trace("Preview not updated")
return
get_images().load_training_preview()
self.display_item = get_images().previewtrain
def display_item_process(self):
""" Display the preview(s) resized as appropriate """
logger.trace("Displaying preview")
sortednames = sorted(list(get_images().previewtrain.keys()))
existing = self.subnotebook_get_titles_ids()
should_update = self.update_preview.get()
for name in sortednames:
if name not in existing.keys():
self.add_child(name)
elif should_update:
tab_id = existing[name]
self.update_child(tab_id, name)
if should_update:
self.update_preview.set(False)
def add_child(self, name):
""" Add the preview canvas child """
logger.debug("Adding child")
preview = PreviewTrainCanvas(self.subnotebook, name)
preview = self.subnotebook_add_page(name, widget=preview)
Tooltip(preview, text=self.helptext, wraplength=200)
self.vars["modified"].set(get_images().previewtrain[name][2])
def update_child(self, tab_id, name):
""" Update the preview canvas """
logger.debug("Updating preview")
if self.vars["modified"].get() != get_images().previewtrain[name][2]:
self.vars["modified"].set(get_images().previewtrain[name][2])
widget = self.subnotebook_page_from_id(tab_id)
widget.reload()
def save_items(self):
""" Open save dialogue and save preview """
location = FileHandler("dir", None).retfile
if not location:
return
for preview in self.subnotebook.children.values():
preview.save_preview(location)
class PreviewTrainCanvas(ttk.Frame): # pylint: disable=too-many-ancestors
""" Canvas to hold a training preview image """
def __init__(self, parent, previewname):
logger.debug("Initializing %s: (previewname: '%s')", self.__class__.__name__, previewname)
ttk.Frame.__init__(self, parent)
self.name = previewname
get_images().resize_image(self.name, None)
self.previewimage = get_images().previewtrain[self.name][1]
self.canvas = tk.Canvas(self, bd=0, highlightthickness=0)
self.canvas.pack(side=tk.TOP, fill=tk.BOTH, expand=True)
self.imgcanvas = self.canvas.create_image(0,
0,
image=self.previewimage,
anchor=tk.NW)
self.bind("<Configure>", self.resize)
logger.debug("Initialized %s:", self.__class__.__name__)
def resize(self, event):
""" Resize the image to fit the frame, maintaining aspect ratio """
logger.trace("Resizing preview image")
framesize = (event.width, event.height)
# Sometimes image is resized before frame is drawn
framesize = None if framesize == (1, 1) else framesize
get_images().resize_image(self.name, framesize)
self.reload()
def reload(self):
""" Reload the preview image """
logger.trace("Reloading preview image")
self.previewimage = get_images().previewtrain[self.name][1]
self.canvas.itemconfig(self.imgcanvas, image=self.previewimage)
def save_preview(self, location):
""" Save the figure to file """
filename = self.name
now = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
filename = os.path.join(location,
"{}_{}.{}".format(filename,
now,
"png"))
get_images().previewtrain[self.name][0].save(filename)
logger.debug("Saved preview to %s", filename)
print("Saved preview to {}".format(filename))
class GraphDisplay(DisplayOptionalPage): # pylint: disable=too-many-ancestors
""" The Graph Tab of the Display section """
def add_options(self):
""" Add the additional options """
self.add_option_refresh()
super().add_options()
def add_option_refresh(self):
""" Add refresh button to refresh graph immediately """
logger.debug("Adding refresh option")
tk_var = get_config().tk_vars["refreshgraph"]
btnrefresh = ttk.Button(self.optsframe,
image=get_images().icons["reset"],
command=lambda: tk_var.set(True))
btnrefresh.pack(padx=2, side=tk.RIGHT)
Tooltip(btnrefresh,
text="Graph updates every 100 iterations. Click to refresh now.",
wraplength=200)
def display_item_set(self):
""" Load the graph(s) if available """
session = get_config().session
if session.initialized and session.logging_disabled:
logger.trace("Logs disabled. Hiding graph")
self.set_info("Graph is disabled as 'no-logs' has been selected")
self.display_item = None
elif session.initialized:
logger.trace("Loading graph")
self.display_item = session
else:
self.display_item = None
def display_item_process(self):
""" Add a single graph to the graph window """
logger.trace("Adding graph")
existing = list(self.subnotebook_get_titles_ids().keys())
for loss_key in self.display_item.loss_keys:
tabname = loss_key.replace("_", " ").title()
if tabname in existing:
continue
data = Calculations(session=get_config().session,
display="loss",
loss_keys=[loss_key],
selections=["raw", "trend"])
self.add_child(tabname, data)
def add_child(self, name, data):
""" Add the graph for the selected keys """
logger.debug("Adding child: %s", name)
graph = TrainingGraph(self.subnotebook, data, "Loss")
graph.build()
graph = self.subnotebook_add_page(name, widget=graph)
Tooltip(graph, text=self.helptext, wraplength=200)
def save_items(self):
""" Open save dialogue and save graphs """
graphlocation = FileHandler("dir", None).retfile
if not graphlocation:
return
for graph in self.subnotebook.children.values():
graph.save_fig(graphlocation)