-
Notifications
You must be signed in to change notification settings - Fork 0
/
tool_add_wheat_control.py
48 lines (36 loc) · 1.35 KB
/
tool_add_wheat_control.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
import os
input_path = 'models/DODA-wheat-ldm.ckpt'
output_path = 'models/control-wheat-ldm.ckpt'
assert os.path.exists(input_path), 'Input model does not exist.'
assert not os.path.exists(output_path), 'Output filename already exists.'
assert os.path.exists(os.path.dirname(output_path)), 'Output path is not valid.'
import torch
from share import *
from cldm.model import create_model
def get_node_name(name, parent_name):
if len(name) <= len(parent_name):
return False, ''
p = name[:len(parent_name)]
if p != parent_name:
return False, ''
return True, name[len(parent_name):]
model = create_model(config_path='./configs/controlnet/DODA_wheat_cldm_kl_4.yaml')
pretrained_weights = torch.load(input_path)
if 'state_dict' in pretrained_weights:
pretrained_weights = pretrained_weights['state_dict']
scratch_dict = model.state_dict()
target_dict = {}
for k in scratch_dict.keys():
is_control, name = get_node_name(k, 'control_')
if is_control:
copy_k = 'model.diffusion_' + name
else:
copy_k = k
if copy_k in pretrained_weights:
target_dict[k] = pretrained_weights[copy_k].clone()
else:
target_dict[k] = scratch_dict[k].clone()
print(f'These weights are newly added: {k}')
model.load_state_dict(target_dict, strict=True)
torch.save(model.state_dict(), output_path)
print('Done.')