mirror of
https://github.com/iperov/DeepFaceLab.git
synced 2025-08-19 04:59:27 -07:00
remove all trailing spaces pt2
This commit is contained in:
camjac251
parent
56f7add24c
commit
b2f287e52e
7 changed files with 112 additions and 112 deletions
|
|
@ -16,7 +16,7 @@ class DeepFakeArchi(nn.ArchiBase):
|
|||
|
||||
|
||||
conv_dtype = tf.float16 if use_fp16 else tf.float32
|
||||
|
||||
|
||||
if mod is None:
|
||||
class Downscale(nn.ModelBase):
|
||||
def __init__(self, in_ch, out_ch, kernel_size=5, *kwargs ):
|
||||
|
|
@ -79,8 +79,8 @@ class DeepFakeArchi(nn.ArchiBase):
|
|||
self.in_ch = in_ch
|
||||
self.e_ch = e_ch
|
||||
super().__init__(**kwargs)
|
||||
|
||||
def on_build(self):
|
||||
|
||||
def on_build(self):
|
||||
self.down1 = DownscaleBlock(self.in_ch, self.e_ch, n_downscales=4, kernel_size=5)
|
||||
|
||||
def forward(self, x):
|
||||
|
|
@ -90,10 +90,10 @@ class DeepFakeArchi(nn.ArchiBase):
|
|||
if use_fp16:
|
||||
x = tf.cast(x, tf.float32)
|
||||
return x
|
||||
|
||||
|
||||
def get_out_res(self, res):
|
||||
return res // (2**4)
|
||||
|
||||
|
||||
def get_out_ch(self):
|
||||
return self.e_ch * 8
|
||||
|
||||
|
|
@ -106,10 +106,10 @@ class DeepFakeArchi(nn.ArchiBase):
|
|||
|
||||
def on_build(self):
|
||||
in_ch, ae_ch, ae_out_ch = self.in_ch, self.ae_ch, self.ae_out_ch
|
||||
|
||||
|
||||
if 'u' in opts:
|
||||
self.dense_norm = nn.DenseNorm()
|
||||
|
||||
|
||||
self.dense1 = nn.Dense( in_ch, ae_ch )
|
||||
self.dense2 = nn.Dense( ae_ch, lowest_dense_res * lowest_dense_res * ae_out_ch )
|
||||
self.upscale1 = Upscale(ae_out_ch, ae_out_ch)
|
||||
|
|
@ -121,7 +121,7 @@ class DeepFakeArchi(nn.ArchiBase):
|
|||
x = self.dense1(x)
|
||||
x = self.dense2(x)
|
||||
x = nn.reshape_4D (x, lowest_dense_res, lowest_dense_res, self.ae_out_ch)
|
||||
|
||||
|
||||
if use_fp16:
|
||||
x = tf.cast(x, tf.float16)
|
||||
x = self.upscale1(x)
|
||||
|
|
@ -134,7 +134,7 @@ class DeepFakeArchi(nn.ArchiBase):
|
|||
return self.ae_out_ch
|
||||
|
||||
class Decoder(nn.ModelBase):
|
||||
def on_build(self, in_ch, d_ch, d_mask_ch):
|
||||
def on_build(self, in_ch, d_ch, d_mask_ch):
|
||||
self.upscale0 = Upscale(in_ch, d_ch*8, kernel_size=3)
|
||||
self.upscale1 = Upscale(d_ch*8, d_ch*4, kernel_size=3)
|
||||
self.upscale2 = Upscale(d_ch*4, d_ch*2, kernel_size=3)
|
||||
|
|
@ -182,13 +182,13 @@ class DeepFakeArchi(nn.ArchiBase):
|
|||
if 'd' in opts:
|
||||
m = self.upscalem3(m)
|
||||
m = tf.nn.sigmoid(self.out_convm(m))
|
||||
|
||||
|
||||
if use_fp16:
|
||||
x = tf.cast(x, tf.float32)
|
||||
x = tf.cast(x, tf.float32)
|
||||
m = tf.cast(m, tf.float32)
|
||||
|
||||
|
||||
return x, m
|
||||
|
||||
|
||||
self.Encoder = Encoder
|
||||
self.Inter = Inter
|
||||
self.Decoder = Decoder
|
||||
|
|
|
|||
|
|
@ -111,7 +111,7 @@ class UNetPatchDiscriminator(nn.ModelBase):
|
|||
for i in range(layers_count-1):
|
||||
st = 1 + (1 if val & (1 << i) !=0 else 0 )
|
||||
layers.append ( [3, st ])
|
||||
sum_st += st
|
||||
sum_st += st
|
||||
|
||||
rf = self.calc_receptive_field_size(layers)
|
||||
|
||||
|
|
@ -132,8 +132,8 @@ class UNetPatchDiscriminator(nn.ModelBase):
|
|||
|
||||
def on_build(self, patch_size, in_ch, base_ch = 16, use_fp16 = False):
|
||||
self.use_fp16 = use_fp16
|
||||
conv_dtype = tf.float16 if use_fp16 else tf.float32
|
||||
|
||||
conv_dtype = tf.float16 if use_fp16 else tf.float32
|
||||
|
||||
class ResidualBlock(nn.ModelBase):
|
||||
def on_build(self, ch, kernel_size=3 ):
|
||||
self.conv1 = nn.Conv2D( ch, ch, kernel_size=kernel_size, padding='SAME', dtype=conv_dtype)
|
||||
|
|
@ -150,7 +150,7 @@ class UNetPatchDiscriminator(nn.ModelBase):
|
|||
self.convs = []
|
||||
self.upconvs = []
|
||||
layers = self.find_archi(patch_size)
|
||||
|
||||
|
||||
level_chs = { i-1:v for i,v in enumerate([ min( base_ch * (2**i), 512 ) for i in range(len(layers)+1)]) }
|
||||
|
||||
self.in_conv = nn.Conv2D( in_ch, level_chs[-1], kernel_size=1, padding='VALID', dtype=conv_dtype)
|
||||
|
|
@ -169,14 +169,14 @@ class UNetPatchDiscriminator(nn.ModelBase):
|
|||
def forward(self, x):
|
||||
if self.use_fp16:
|
||||
x = tf.cast(x, tf.float16)
|
||||
|
||||
|
||||
x = tf.nn.leaky_relu( self.in_conv(x), 0.2 )
|
||||
|
||||
encs = []
|
||||
for conv in self.convs:
|
||||
encs.insert(0, x)
|
||||
x = tf.nn.leaky_relu( conv(x), 0.2 )
|
||||
|
||||
|
||||
center_out, x = self.center_out(x), tf.nn.leaky_relu( self.center_conv(x), 0.2 )
|
||||
|
||||
for i, (upconv, enc) in enumerate(zip(self.upconvs, encs)):
|
||||
|
|
@ -184,7 +184,7 @@ class UNetPatchDiscriminator(nn.ModelBase):
|
|||
x = tf.concat( [enc, x], axis=nn.conv2d_ch_axis)
|
||||
|
||||
x = self.out_conv(x)
|
||||
|
||||
|
||||
if self.use_fp16:
|
||||
center_out = tf.cast(center_out, tf.float32)
|
||||
x = tf.cast(x, tf.float32)
|
||||
|
|
|
|||
42
main.py
42
main.py
|
|
@ -23,7 +23,7 @@ if __name__ == "__main__":
|
|||
setattr(namespace, self.dest, os.path.abspath(os.path.expanduser(values)))
|
||||
|
||||
exit_code = 0
|
||||
|
||||
|
||||
parser = argparse.ArgumentParser()
|
||||
subparsers = parser.add_subparsers()
|
||||
|
||||
|
|
@ -52,9 +52,9 @@ if __name__ == "__main__":
|
|||
p.add_argument('--output-debug', action="store_true", dest="output_debug", default=None, help="Writes debug images to <output-dir>_debug\ directory.")
|
||||
p.add_argument('--no-output-debug', action="store_false", dest="output_debug", default=None, help="Don't writes debug images to <output-dir>_debug\ directory.")
|
||||
p.add_argument('--face-type', dest="face_type", choices=['half_face', 'full_face', 'whole_face', 'head', 'mark_only'], default=None)
|
||||
p.add_argument('--max-faces-from-image', type=int, dest="max_faces_from_image", default=None, help="Max faces from image.")
|
||||
p.add_argument('--max-faces-from-image', type=int, dest="max_faces_from_image", default=None, help="Max faces from image.")
|
||||
p.add_argument('--image-size', type=int, dest="image_size", default=None, help="Output image size.")
|
||||
p.add_argument('--jpeg-quality', type=int, dest="jpeg_quality", default=None, help="Jpeg quality.")
|
||||
p.add_argument('--jpeg-quality', type=int, dest="jpeg_quality", default=None, help="Jpeg quality.")
|
||||
p.add_argument('--manual-fix', action="store_true", dest="manual_fix", default=False, help="Enables manual extract only frames where faces were not recognized.")
|
||||
p.add_argument('--manual-output-debug-fix', action="store_true", dest="manual_output_debug_fix", default=False, help="Performs manual reextract input-dir frames which were deleted from [output_dir]_debug\ dir.")
|
||||
p.add_argument('--manual-window-size', type=int, dest="manual_window_size", default=1368, help="Manual fix window size. Default: 1368.")
|
||||
|
|
@ -144,10 +144,10 @@ if __name__ == "__main__":
|
|||
p.add_argument('--cpu-only', action="store_true", dest="cpu_only", default=False, help="Train on CPU.")
|
||||
p.add_argument('--force-gpu-idxs', dest="force_gpu_idxs", default=None, help="Force to choose GPU indexes separated by comma.")
|
||||
p.add_argument('--silent-start', action="store_true", dest="silent_start", default=False, help="Silent start. Automatically chooses Best GPU and last used model.")
|
||||
|
||||
|
||||
p.add_argument('--execute-program', dest="execute_program", default=[], action='append', nargs='+')
|
||||
p.set_defaults (func=process_train)
|
||||
|
||||
|
||||
def process_exportdfm(arguments):
|
||||
osex.set_process_lowest_prio()
|
||||
from mainscripts import ExportDFM
|
||||
|
|
@ -261,8 +261,8 @@ if __name__ == "__main__":
|
|||
p.add_argument('--force-gpu-idxs', dest="force_gpu_idxs", default=None, help="Force to choose GPU indexes separated by comma.")
|
||||
|
||||
p.set_defaults(func=process_faceset_enhancer)
|
||||
|
||||
|
||||
|
||||
|
||||
p = facesettool_parser.add_parser ("resize", help="Resize DFL faceset.")
|
||||
p.add_argument('--input-dir', required=True, action=fixPathAction, dest="input_dir", help="Input directory of aligned faces.")
|
||||
|
||||
|
|
@ -271,7 +271,7 @@ if __name__ == "__main__":
|
|||
from mainscripts import FacesetResizer
|
||||
FacesetResizer.process_folder ( Path(arguments.input_dir) )
|
||||
p.set_defaults(func=process_faceset_resizer)
|
||||
|
||||
|
||||
def process_dev_test(arguments):
|
||||
osex.set_process_lowest_prio()
|
||||
from mainscripts import dev_misc
|
||||
|
|
@ -280,10 +280,10 @@ if __name__ == "__main__":
|
|||
p = subparsers.add_parser( "dev_test", help="")
|
||||
p.add_argument('--input-dir', required=True, action=fixPathAction, dest="input_dir")
|
||||
p.set_defaults (func=process_dev_test)
|
||||
|
||||
|
||||
# ========== XSeg
|
||||
xseg_parser = subparsers.add_parser( "xseg", help="XSeg tools.").add_subparsers()
|
||||
|
||||
|
||||
p = xseg_parser.add_parser( "editor", help="XSeg editor.")
|
||||
|
||||
def process_xsegeditor(arguments):
|
||||
|
|
@ -291,11 +291,11 @@ if __name__ == "__main__":
|
|||
from XSegEditor import XSegEditor
|
||||
global exit_code
|
||||
exit_code = XSegEditor.start (Path(arguments.input_dir))
|
||||
|
||||
|
||||
p.add_argument('--input-dir', required=True, action=fixPathAction, dest="input_dir")
|
||||
|
||||
p.set_defaults (func=process_xsegeditor)
|
||||
|
||||
|
||||
p = xseg_parser.add_parser( "apply", help="Apply trained XSeg model to the extracted faces.")
|
||||
|
||||
def process_xsegapply(arguments):
|
||||
|
|
@ -305,8 +305,8 @@ if __name__ == "__main__":
|
|||
p.add_argument('--input-dir', required=True, action=fixPathAction, dest="input_dir")
|
||||
p.add_argument('--model-dir', required=True, action=fixPathAction, dest="model_dir")
|
||||
p.set_defaults (func=process_xsegapply)
|
||||
|
||||
|
||||
|
||||
|
||||
p = xseg_parser.add_parser( "remove", help="Remove applied XSeg masks from the extracted faces.")
|
||||
def process_xsegremove(arguments):
|
||||
osex.set_process_lowest_prio()
|
||||
|
|
@ -314,8 +314,8 @@ if __name__ == "__main__":
|
|||
XSegUtil.remove_xseg (Path(arguments.input_dir) )
|
||||
p.add_argument('--input-dir', required=True, action=fixPathAction, dest="input_dir")
|
||||
p.set_defaults (func=process_xsegremove)
|
||||
|
||||
|
||||
|
||||
|
||||
p = xseg_parser.add_parser( "remove_labels", help="Remove XSeg labels from the extracted faces.")
|
||||
def process_xsegremovelabels(arguments):
|
||||
osex.set_process_lowest_prio()
|
||||
|
|
@ -323,8 +323,8 @@ if __name__ == "__main__":
|
|||
XSegUtil.remove_xseg_labels (Path(arguments.input_dir) )
|
||||
p.add_argument('--input-dir', required=True, action=fixPathAction, dest="input_dir")
|
||||
p.set_defaults (func=process_xsegremovelabels)
|
||||
|
||||
|
||||
|
||||
|
||||
p = xseg_parser.add_parser( "fetch", help="Copies faces containing XSeg polygons in <input_dir>_xseg dir.")
|
||||
|
||||
def process_xsegfetch(arguments):
|
||||
|
|
@ -333,7 +333,7 @@ if __name__ == "__main__":
|
|||
XSegUtil.fetch_xseg (Path(arguments.input_dir) )
|
||||
p.add_argument('--input-dir', required=True, action=fixPathAction, dest="input_dir")
|
||||
p.set_defaults (func=process_xsegfetch)
|
||||
|
||||
|
||||
def bad_args(arguments):
|
||||
parser.print_help()
|
||||
exit(0)
|
||||
|
|
@ -344,9 +344,9 @@ if __name__ == "__main__":
|
|||
|
||||
if exit_code == 0:
|
||||
print ("Done.")
|
||||
|
||||
|
||||
exit(exit_code)
|
||||
|
||||
|
||||
'''
|
||||
import code
|
||||
code.interact(local=dict(globals(), **locals()))
|
||||
|
|
|
|||
|
|
@ -19,4 +19,4 @@ def main(model_class_name, saved_models_path):
|
|||
is_exporting=True,
|
||||
saved_models_path=saved_models_path,
|
||||
cpu_only=True)
|
||||
model.export_dfm ()
|
||||
model.export_dfm ()
|
||||
|
|
|
|||
|
|
@ -79,79 +79,79 @@ class FacesetResizerSubprocessor(Subprocessor):
|
|||
h,w = img.shape[:2]
|
||||
if h != w:
|
||||
raise Exception(f'w != h in {filepath}')
|
||||
|
||||
|
||||
image_size = self.image_size
|
||||
face_type = self.face_type
|
||||
output_filepath = self.output_dirpath / filepath.name
|
||||
|
||||
|
||||
if face_type is not None:
|
||||
lmrks = dflimg.get_landmarks()
|
||||
mat = LandmarksProcessor.get_transform_mat(lmrks, image_size, face_type)
|
||||
|
||||
|
||||
img = cv2.warpAffine(img, mat, (image_size, image_size), flags=cv2.INTER_LANCZOS4 )
|
||||
img = np.clip(img, 0, 255).astype(np.uint8)
|
||||
|
||||
|
||||
cv2_imwrite ( str(output_filepath), img, [int(cv2.IMWRITE_JPEG_QUALITY), 100] )
|
||||
|
||||
dfl_dict = dflimg.get_dict()
|
||||
dflimg = DFLIMG.load (output_filepath)
|
||||
dflimg.set_dict(dfl_dict)
|
||||
|
||||
|
||||
xseg_mask = dflimg.get_xseg_mask()
|
||||
if xseg_mask is not None:
|
||||
xseg_res = 256
|
||||
|
||||
|
||||
xseg_lmrks = lmrks.copy()
|
||||
xseg_lmrks *= (xseg_res / w)
|
||||
xseg_mat = LandmarksProcessor.get_transform_mat(xseg_lmrks, xseg_res, face_type)
|
||||
|
||||
|
||||
xseg_mask = cv2.warpAffine(xseg_mask, xseg_mat, (xseg_res, xseg_res), flags=cv2.INTER_LANCZOS4 )
|
||||
xseg_mask[xseg_mask < 0.5] = 0
|
||||
xseg_mask[xseg_mask >= 0.5] = 1
|
||||
|
||||
dflimg.set_xseg_mask(xseg_mask)
|
||||
|
||||
|
||||
seg_ie_polys = dflimg.get_seg_ie_polys()
|
||||
|
||||
|
||||
for poly in seg_ie_polys.get_polys():
|
||||
poly_pts = poly.get_pts()
|
||||
poly_pts = LandmarksProcessor.transform_points(poly_pts, mat)
|
||||
poly.set_points(poly_pts)
|
||||
|
||||
|
||||
dflimg.set_seg_ie_polys(seg_ie_polys)
|
||||
|
||||
|
||||
lmrks = LandmarksProcessor.transform_points(lmrks, mat)
|
||||
dflimg.set_landmarks(lmrks)
|
||||
|
||||
|
||||
image_to_face_mat = dflimg.get_image_to_face_mat()
|
||||
if image_to_face_mat is not None:
|
||||
image_to_face_mat = LandmarksProcessor.get_transform_mat ( dflimg.get_source_landmarks(), image_size, face_type )
|
||||
dflimg.set_image_to_face_mat(image_to_face_mat)
|
||||
dflimg.set_face_type( FaceType.toString(face_type) )
|
||||
dflimg.save()
|
||||
|
||||
|
||||
else:
|
||||
dfl_dict = dflimg.get_dict()
|
||||
|
||||
|
||||
scale = w / image_size
|
||||
|
||||
img = cv2.resize(img, (image_size, image_size), interpolation=cv2.INTER_LANCZOS4)
|
||||
|
||||
|
||||
img = cv2.resize(img, (image_size, image_size), interpolation=cv2.INTER_LANCZOS4)
|
||||
|
||||
cv2_imwrite ( str(output_filepath), img, [int(cv2.IMWRITE_JPEG_QUALITY), 100] )
|
||||
|
||||
dflimg = DFLIMG.load (output_filepath)
|
||||
dflimg.set_dict(dfl_dict)
|
||||
|
||||
lmrks = dflimg.get_landmarks()
|
||||
|
||||
lmrks = dflimg.get_landmarks()
|
||||
lmrks /= scale
|
||||
dflimg.set_landmarks(lmrks)
|
||||
|
||||
|
||||
seg_ie_polys = dflimg.get_seg_ie_polys()
|
||||
seg_ie_polys.mult_points( 1.0 / scale)
|
||||
dflimg.set_seg_ie_polys(seg_ie_polys)
|
||||
|
||||
|
||||
image_to_face_mat = dflimg.get_image_to_face_mat()
|
||||
|
||||
|
||||
if image_to_face_mat is not None:
|
||||
face_type = FaceType.fromString ( dflimg.get_face_type() )
|
||||
image_to_face_mat = LandmarksProcessor.get_transform_mat ( dflimg.get_source_landmarks(), image_size, face_type )
|
||||
|
|
@ -165,9 +165,9 @@ class FacesetResizerSubprocessor(Subprocessor):
|
|||
return (0, filepath, None)
|
||||
|
||||
def process_folder ( dirpath):
|
||||
|
||||
|
||||
image_size = io.input_int(f"New image size", 512, valid_range=[128,2048])
|
||||
|
||||
|
||||
face_type = io.input_str ("Change face type", 'same', ['h','mf','f','wf','head','same']).lower()
|
||||
if face_type == 'same':
|
||||
face_type = None
|
||||
|
|
@ -177,7 +177,7 @@ def process_folder ( dirpath):
|
|||
'f' : FaceType.FULL,
|
||||
'wf' : FaceType.WHOLE_FACE,
|
||||
'head' : FaceType.HEAD}[face_type]
|
||||
|
||||
|
||||
|
||||
output_dirpath = dirpath.parent / (dirpath.name + '_resized')
|
||||
output_dirpath.mkdir (exist_ok=True, parents=True)
|
||||
|
|
|
|||
|
|
@ -42,7 +42,7 @@ def trainerThread (s2c, c2s, e,
|
|||
|
||||
if not saved_models_path.exists():
|
||||
saved_models_path.mkdir(exist_ok=True, parents=True)
|
||||
|
||||
|
||||
model = models.import_model(model_class_name)(
|
||||
is_training=True,
|
||||
saved_models_path=saved_models_path,
|
||||
|
|
@ -67,10 +67,10 @@ def trainerThread (s2c, c2s, e,
|
|||
io.log_info ("Saving....", end='\r')
|
||||
model.save()
|
||||
shared_state['after_save'] = True
|
||||
|
||||
|
||||
def model_backup():
|
||||
if not debug and not is_reached_goal:
|
||||
model.create_backup()
|
||||
model.create_backup()
|
||||
|
||||
def send_preview():
|
||||
if not debug:
|
||||
|
|
@ -119,7 +119,7 @@ def trainerThread (s2c, c2s, e,
|
|||
io.log_info("")
|
||||
io.log_info("Trying to do the first iteration. If an error occurs, reduce the model parameters.")
|
||||
io.log_info("")
|
||||
|
||||
|
||||
if sys.platform[0:3] == 'win':
|
||||
io.log_info("!!!")
|
||||
io.log_info("Windows 10 users IMPORTANT notice. You should set this setting in order to work correctly.")
|
||||
|
|
@ -137,7 +137,7 @@ def trainerThread (s2c, c2s, e,
|
|||
|
||||
if shared_state['after_save']:
|
||||
shared_state['after_save'] = False
|
||||
|
||||
|
||||
mean_loss = np.mean ( loss_history[save_iter:iter], axis=0)
|
||||
|
||||
for loss_value in mean_loss:
|
||||
|
|
|
|||
|
|
@ -29,7 +29,7 @@ class SAEHDModel(ModelBase):
|
|||
yn_str = {True:'y',False:'n'}
|
||||
min_res = 64
|
||||
max_res = 640
|
||||
|
||||
|
||||
#default_usefp16 = self.options['use_fp16'] = self.load_or_def_option('use_fp16', False)
|
||||
default_resolution = self.options['resolution'] = self.load_or_def_option('resolution', 128)
|
||||
default_face_type = self.options['face_type'] = self.load_or_def_option('face_type', 'f')
|
||||
|
|
@ -70,14 +70,14 @@ class SAEHDModel(ModelBase):
|
|||
self.ask_random_dst_flip()
|
||||
self.ask_batch_size(suggest_batch_size)
|
||||
#self.options['use_fp16'] = io.input_bool ("Use fp16", default_usefp16, help_message='Increases training/inference speed, reduces model size. Model may crash. Enable it after 1-5k iters.')
|
||||
|
||||
|
||||
if self.is_first_run():
|
||||
resolution = io.input_int("Resolution", default_resolution, add_info="64-640", help_message="More resolution requires more VRAM and time to train. Value will be adjusted to multiple of 16 and 32 for -d archi.")
|
||||
resolution = np.clip ( (resolution // 16) * 16, min_res, max_res)
|
||||
self.options['resolution'] = resolution
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
self.options['face_type'] = io.input_str ("Face type", default_face_type, ['h','mf','f','wf','head'], help_message="Half / mid face / full face / whole face / head. Half face has better resolution, but covers less area of cheeks. Mid face is 30% wider than half face. 'Whole face' covers full area of face include forehead. 'head' covers full head, but requires XSeg for src and dst faceset.").lower()
|
||||
|
||||
while True:
|
||||
|
|
@ -138,11 +138,11 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
|
||||
self.options['eyes_mouth_prio'] = io.input_bool ("Eyes and mouth priority", default_eyes_mouth_prio, help_message='Helps to fix eye problems during training like "alien eyes" and wrong eyes direction. Also makes the detail of the teeth higher.')
|
||||
self.options['uniform_yaw'] = io.input_bool ("Uniform yaw distribution of samples", default_uniform_yaw, help_message='Helps to fix blurry side faces due to small amount of them in the faceset.')
|
||||
|
||||
|
||||
default_gan_power = self.options['gan_power'] = self.load_or_def_option('gan_power', 0.0)
|
||||
default_gan_patch_size = self.options['gan_patch_size'] = self.load_or_def_option('gan_patch_size', self.options['resolution'] // 8)
|
||||
default_gan_dims = self.options['gan_dims'] = self.load_or_def_option('gan_dims', 16)
|
||||
|
||||
|
||||
if self.is_first_run() or ask_override:
|
||||
self.options['models_opt_on_gpu'] = io.input_bool ("Place models and optimizer on GPU", default_models_opt_on_gpu, help_message="When you train on one GPU, by default model and optimizer weights are placed on GPU to accelerate the process. You can place they on CPU to free up extra VRAM, thus set bigger dimensions.")
|
||||
|
||||
|
|
@ -153,14 +153,14 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
self.options['random_warp'] = io.input_bool ("Enable random warp of samples", default_random_warp, help_message="Random warp is required to generalize facial expressions of both faces. When the face is trained enough, you can disable it to get extra sharpness and reduce subpixel shake for less amount of iterations.")
|
||||
|
||||
self.options['gan_power'] = np.clip ( io.input_number ("GAN power", default_gan_power, add_info="0.0 .. 5.0", help_message="Forces the neural network to learn small details of the face. Enable it only when the face is trained enough with lr_dropout(on) and random_warp(off), and don't disable. The higher the value, the higher the chances of artifacts. Typical fine value is 0.1"), 0.0, 5.0 )
|
||||
|
||||
if self.options['gan_power'] != 0.0:
|
||||
|
||||
if self.options['gan_power'] != 0.0:
|
||||
gan_patch_size = np.clip ( io.input_int("GAN patch size", default_gan_patch_size, add_info="3-640", help_message="The higher patch size, the higher the quality, the more VRAM is required. You can get sharper edges even at the lowest setting. Typical fine value is resolution / 8." ), 3, 640 )
|
||||
self.options['gan_patch_size'] = gan_patch_size
|
||||
|
||||
|
||||
gan_dims = np.clip ( io.input_int("GAN dimensions", default_gan_dims, add_info="4-512", help_message="The dimensions of the GAN network. The higher dimensions, the more VRAM is required. You can get sharper edges even at the lowest setting. Typical fine value is 16." ), 4, 512 )
|
||||
self.options['gan_dims'] = gan_dims
|
||||
|
||||
|
||||
if 'df' in self.options['archi']:
|
||||
self.options['true_face_power'] = np.clip ( io.input_number ("'True face' power.", default_true_face_power, add_info="0.0000 .. 1.0", help_message="Experimental option. Discriminates result face to be more like src face. Higher value - stronger discrimination. Typical value is 0.01 . Comparison - https://i.imgur.com/czScS9q.png"), 0.0, 1.0 )
|
||||
else:
|
||||
|
|
@ -176,7 +176,7 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
|
||||
if self.options['pretrain'] and self.get_pretraining_data_path() is None:
|
||||
raise Exception("pretraining_data_path is not defined")
|
||||
|
||||
|
||||
self.gan_model_changed = (default_gan_patch_size != self.options['gan_patch_size']) or (default_gan_dims != self.options['gan_dims'])
|
||||
|
||||
self.pretrain_just_disabled = (default_pretrain == True and self.options['pretrain'] == False)
|
||||
|
|
@ -198,7 +198,7 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
|
||||
if 'eyes_prio' in self.options:
|
||||
self.options.pop('eyes_prio')
|
||||
|
||||
|
||||
eyes_mouth_prio = self.options['eyes_mouth_prio']
|
||||
|
||||
archi_split = self.options['archi'].split('-')
|
||||
|
|
@ -207,7 +207,7 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
archi_type, archi_opts = archi_split
|
||||
elif len(archi_split) == 1:
|
||||
archi_type, archi_opts = archi_split[0], None
|
||||
|
||||
|
||||
self.archi_type = archi_type
|
||||
|
||||
ae_dims = self.options['ae_dims']
|
||||
|
|
@ -220,12 +220,12 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
|
||||
adabelief = self.options['adabelief']
|
||||
use_fp16 = False#self.options['use_fp16']
|
||||
|
||||
|
||||
self.gan_power = gan_power = 0.0 if self.pretrain else self.options['gan_power']
|
||||
random_warp = False if self.pretrain else self.options['random_warp']
|
||||
random_src_flip = self.random_src_flip if not self.pretrain else True
|
||||
random_dst_flip = self.random_dst_flip if not self.pretrain else True
|
||||
|
||||
|
||||
if self.pretrain:
|
||||
self.options_show_override['gan_power'] = 0.0
|
||||
self.options_show_override['random_warp'] = False
|
||||
|
|
@ -238,8 +238,8 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
ct_mode = self.options['ct_mode']
|
||||
if ct_mode == 'none':
|
||||
ct_mode = None
|
||||
|
||||
|
||||
|
||||
|
||||
models_opt_on_gpu = False if len(devices) == 0 else self.options['models_opt_on_gpu']
|
||||
models_opt_device = nn.tf_default_device_name if models_opt_on_gpu and self.is_training else '/CPU:0'
|
||||
optimizer_vars_on_cpu = models_opt_device=='/CPU:0'
|
||||
|
|
@ -353,7 +353,7 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
gpu_G_loss_gvs = []
|
||||
gpu_D_code_loss_gvs = []
|
||||
gpu_D_src_dst_loss_gvs = []
|
||||
|
||||
|
||||
for gpu_id in range(gpu_count):
|
||||
with tf.device( f'/{devices[gpu_id].tf_dev_type}:{gpu_id}' if len(devices) != 0 else f'/CPU:0' ):
|
||||
with tf.device(f'/CPU:0'):
|
||||
|
|
@ -405,7 +405,7 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
gpu_target_dstm_style_blur = gpu_target_dstm_blur #default style mask is 0.5 on boundary
|
||||
gpu_target_dstm_blur = tf.clip_by_value(gpu_target_dstm_blur, 0, 0.5) * 2
|
||||
|
||||
gpu_target_dst_masked = gpu_target_dst*gpu_target_dstm_blur
|
||||
gpu_target_dst_masked = gpu_target_dst*gpu_target_dstm_blur
|
||||
gpu_target_dst_style_masked = gpu_target_dst*gpu_target_dstm_style_blur
|
||||
gpu_target_dst_style_anti_masked = gpu_target_dst*(1.0 - gpu_target_dstm_style_blur)
|
||||
|
||||
|
|
@ -500,14 +500,14 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
|
||||
gpu_G_loss += gan_power*(DLoss(gpu_pred_src_src_d_ones, gpu_pred_src_src_d) + \
|
||||
DLoss(gpu_pred_src_src_d2_ones, gpu_pred_src_src_d2))
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
if masked_training:
|
||||
# Minimal src-src-bg rec with total_variation_mse to suppress random bright dots from gan
|
||||
gpu_G_loss += 0.000001*nn.total_variation_mse(gpu_pred_src_src)
|
||||
gpu_G_loss += 0.02*tf.reduce_mean(tf.square(gpu_pred_src_src_anti_masked-gpu_target_src_anti_masked),axis=[1,2,3] )
|
||||
|
||||
|
||||
gpu_G_loss_gvs += [ nn.gradients ( gpu_G_loss, self.src_dst_trainable_weights ) ]
|
||||
|
||||
|
||||
|
|
@ -617,10 +617,10 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
|
||||
if do_init:
|
||||
model.init_weights()
|
||||
|
||||
|
||||
|
||||
|
||||
###############
|
||||
|
||||
|
||||
# initializing sample generators
|
||||
if self.is_training:
|
||||
training_data_src_path = self.training_data_src_path if not self.pretrain else self.get_pretraining_data_path()
|
||||
|
|
@ -661,20 +661,20 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
|
||||
if self.pretrain_just_disabled:
|
||||
self.update_sample_for_preview(force_new=True)
|
||||
|
||||
|
||||
def export_dfm (self):
|
||||
output_path=self.get_strpath_storage_for_file('model.dfm')
|
||||
|
||||
|
||||
io.log_info(f'Dumping .dfm to {output_path}')
|
||||
|
||||
|
||||
tf = nn.tf
|
||||
nn.set_data_format('NCHW')
|
||||
|
||||
|
||||
with tf.device (nn.tf_default_device_name):
|
||||
warped_dst = tf.placeholder (nn.floatx, (None, self.resolution, self.resolution, 3), name='in_face')
|
||||
warped_dst = tf.transpose(warped_dst, (0,3,1,2))
|
||||
|
||||
|
||||
|
||||
|
||||
if 'df' in self.archi_type:
|
||||
gpu_dst_code = self.inter(self.encoder(warped_dst))
|
||||
gpu_pred_src_dst, gpu_pred_src_dstm = self.decoder_src(gpu_dst_code)
|
||||
|
|
@ -689,21 +689,21 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
|
||||
gpu_pred_src_dst, gpu_pred_src_dstm = self.decoder(gpu_src_dst_code)
|
||||
_, gpu_pred_dst_dstm = self.decoder(gpu_dst_code)
|
||||
|
||||
|
||||
gpu_pred_src_dst = tf.transpose(gpu_pred_src_dst, (0,2,3,1))
|
||||
gpu_pred_dst_dstm = tf.transpose(gpu_pred_dst_dstm, (0,2,3,1))
|
||||
gpu_pred_src_dstm = tf.transpose(gpu_pred_src_dstm, (0,2,3,1))
|
||||
|
||||
tf.identity(gpu_pred_dst_dstm, name='out_face_mask')
|
||||
tf.identity(gpu_pred_src_dst, name='out_celeb_face')
|
||||
tf.identity(gpu_pred_src_dstm, name='out_celeb_face_mask')
|
||||
|
||||
tf.identity(gpu_pred_src_dstm, name='out_celeb_face_mask')
|
||||
|
||||
output_graph_def = tf.graph_util.convert_variables_to_constants(
|
||||
nn.tf_sess,
|
||||
tf.get_default_graph().as_graph_def(),
|
||||
nn.tf_sess,
|
||||
tf.get_default_graph().as_graph_def(),
|
||||
['out_face_mask','out_celeb_face','out_celeb_face_mask']
|
||||
)
|
||||
|
||||
)
|
||||
|
||||
import tf2onnx
|
||||
with tf.device("/CPU:0"):
|
||||
model_proto, _ = tf2onnx.convert._convert_common(
|
||||
|
|
@ -713,7 +713,7 @@ Examples: df, liae, df-d, df-ud, liae-ud, ...
|
|||
output_names=['out_face_mask:0','out_celeb_face:0','out_celeb_face_mask:0'],
|
||||
opset=13,
|
||||
output_path=output_path)
|
||||
|
||||
|
||||
#override
|
||||
def get_model_filename_list(self):
|
||||
return self.model_filename_list
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue