|
|
|
|
@ -140,7 +140,7 @@ def run_pnginfo(image):
|
|
|
|
|
return '', geninfo, info
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def run_modelmerger(modelname_0, modelname_1, interp_method, interp_amount):
|
|
|
|
|
def run_modelmerger(from_model_name, to_model_name, interp_method, interp_amount):
|
|
|
|
|
# Linear interpolation (https://en.wikipedia.org/wiki/Linear_interpolation)
|
|
|
|
|
def weighted_sum(theta0, theta1, alpha):
|
|
|
|
|
return ((1 - alpha) * theta0) + (alpha * theta1)
|
|
|
|
|
@ -150,23 +150,23 @@ def run_modelmerger(modelname_0, modelname_1, interp_method, interp_amount):
|
|
|
|
|
alpha = alpha * alpha * (3 - (2 * alpha))
|
|
|
|
|
return theta0 + ((theta1 - theta0) * alpha)
|
|
|
|
|
|
|
|
|
|
if os.path.exists(modelname_0):
|
|
|
|
|
model0_filename = modelname_0
|
|
|
|
|
modelname_0 = os.path.splitext(os.path.basename(modelname_0))[0]
|
|
|
|
|
if os.path.exists(to_model_name):
|
|
|
|
|
to_model_filename = to_model_name
|
|
|
|
|
to_model_name = os.path.splitext(os.path.basename(to_model_name))[0]
|
|
|
|
|
else:
|
|
|
|
|
model0_filename = 'models/' + modelname_0 + '.ckpt'
|
|
|
|
|
to_model_filename = 'models/' + to_model_name + '.ckpt'
|
|
|
|
|
|
|
|
|
|
if os.path.exists(modelname_1):
|
|
|
|
|
model1_filename = modelname_1
|
|
|
|
|
modelname_1 = os.path.splitext(os.path.basename(modelname_1))[0]
|
|
|
|
|
if os.path.exists(from_model_name):
|
|
|
|
|
from_model_filename = from_model_name
|
|
|
|
|
from_model_name = os.path.splitext(os.path.basename(from_model_name))[0]
|
|
|
|
|
else:
|
|
|
|
|
model1_filename = 'models/' + modelname_1 + '.ckpt'
|
|
|
|
|
from_model_filename = 'models/' + from_model_name + '.ckpt'
|
|
|
|
|
|
|
|
|
|
print(f"Loading {model0_filename}...")
|
|
|
|
|
model_0 = torch.load(model0_filename, map_location='cpu')
|
|
|
|
|
print(f"Loading {to_model_filename}...")
|
|
|
|
|
model_0 = torch.load(to_model_filename, map_location='cpu')
|
|
|
|
|
|
|
|
|
|
print(f"Loading {model1_filename}...")
|
|
|
|
|
model_1 = torch.load(model1_filename, map_location='cpu')
|
|
|
|
|
print(f"Loading {from_model_filename}...")
|
|
|
|
|
model_1 = torch.load(from_model_filename, map_location='cpu')
|
|
|
|
|
|
|
|
|
|
theta_0 = model_0['state_dict']
|
|
|
|
|
theta_1 = model_1['state_dict']
|
|
|
|
|
@ -186,7 +186,7 @@ def run_modelmerger(modelname_0, modelname_1, interp_method, interp_amount):
|
|
|
|
|
if 'model' in key and key not in theta_0:
|
|
|
|
|
theta_0[key] = theta_1[key]
|
|
|
|
|
|
|
|
|
|
output_modelname = 'models/' + modelname_0 + '-' + modelname_1 + '-' + interp_method.replace(" ", "_") + '-' + str(interp_amount) + '-merged.ckpt'
|
|
|
|
|
output_modelname = 'models/' + from_model_name + str(interp_amount) + '-' + to_model_name + str(float(1.0) - interp_amount) + '-' + interp_method.replace(" ", "_") + '-' + '-merged.ckpt'
|
|
|
|
|
print(f"Saving to {output_modelname}...")
|
|
|
|
|
torch.save(model_0, output_modelname)
|
|
|
|
|
|
|
|
|
|
|