Spaces:
Running
Running
working on arc
Browse files- app.py +289 -10
- ldm/models/__pycache__/autoencoder.cpython-38.pyc +0 -0
- ldm/models/diffusion/__pycache__/__init__.cpython-38.pyc +0 -0
- ldm/models/diffusion/__pycache__/ddim.cpython-38.pyc +0 -0
- ldm/models/diffusion/__pycache__/ddpm.cpython-38.pyc +0 -0
- ldm/models/diffusion/__pycache__/plms.cpython-38.pyc +0 -0
- ldm/models/disentanglement/__pycache__/iterative_normalization.cpython-38.pyc +0 -0
- ldm/modules/diffusionmodules/__pycache__/__init__.cpython-38.pyc +0 -0
- ldm/modules/diffusionmodules/__pycache__/model.cpython-38.pyc +0 -0
- ldm/modules/diffusionmodules/__pycache__/openaimodel.cpython-38.pyc +0 -0
- ldm/modules/diffusionmodules/__pycache__/util.cpython-38.pyc +0 -0
- ldm/modules/distributions/__pycache__/__init__.cpython-38.pyc +0 -0
- ldm/modules/distributions/__pycache__/distributions.cpython-38.pyc +0 -0
- ldm/modules/encoders/__pycache__/__init__.cpython-38.pyc +0 -0
- ldm/modules/encoders/__pycache__/modules.cpython-38.pyc +0 -0
- phylogeny_tree.jpg +0 -0
- phylogeny_tree.pdf +0 -0
- phylogeny_tree.png +0 -0
- sample_level_encoding.py +14 -0
app.py
CHANGED
@@ -2,17 +2,296 @@ import torch
|
|
2 |
import gradio as gr
|
3 |
|
4 |
|
5 |
-
|
6 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
|
14 |
-
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
|
16 |
|
17 |
if __name__ == "__main__":
|
18 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
import gradio as gr
|
3 |
|
4 |
|
5 |
+
import argparse, os, sys, glob
|
6 |
+
import torch
|
7 |
+
import pickle
|
8 |
+
import numpy as np
|
9 |
+
from omegaconf import OmegaConf
|
10 |
+
from PIL import Image
|
11 |
+
from tqdm import tqdm, trange
|
12 |
+
from einops import rearrange
|
13 |
+
from torchvision.utils import make_grid
|
14 |
+
|
15 |
+
from ldm.util import instantiate_from_config
|
16 |
+
from ldm.models.diffusion.ddim import DDIMSampler
|
17 |
+
from ldm.models.diffusion.plms import PLMSSampler
|
18 |
+
|
19 |
+
|
20 |
+
def load_model_from_config(config, ckpt, verbose=False):
|
21 |
+
print(f"Loading model from {ckpt}")
|
22 |
+
# pl_sd = torch.load(ckpt, map_location="cpu")
|
23 |
+
pl_sd = torch.load(ckpt)#, map_location="cpu")
|
24 |
+
sd = pl_sd["state_dict"]
|
25 |
+
model = instantiate_from_config(config.model)
|
26 |
+
m, u = model.load_state_dict(sd, strict=False)
|
27 |
+
if len(m) > 0 and verbose:
|
28 |
+
print("missing keys:")
|
29 |
+
print(m)
|
30 |
+
if len(u) > 0 and verbose:
|
31 |
+
print("unexpected keys:")
|
32 |
+
print(u)
|
33 |
+
|
34 |
+
model.cuda()
|
35 |
+
model.eval()
|
36 |
+
return model
|
37 |
+
|
38 |
+
|
39 |
+
def masking_embed(embedding, levels=1):
|
40 |
+
"""
|
41 |
+
size of embedding - nx1xd, n: number of samples, d - 512
|
42 |
+
replacing the last 128*levels from the embedding
|
43 |
+
"""
|
44 |
+
replace_size = 128*levels
|
45 |
+
random_noise = torch.randn(embedding.shape[0], embedding.shape[1], replace_size)
|
46 |
+
embedding[:, :, -replace_size:] = random_noise
|
47 |
+
return embedding
|
48 |
+
|
49 |
+
def generate_image(fish_name, masking_level_input,
|
50 |
+
swap_fish_name, swap_level_input):
|
51 |
+
|
52 |
+
fish_name = fish_name.lower()
|
53 |
+
|
54 |
+
ckpt_path = '/globalscratch/mridul/ldm/final_runs_eccv/fishes/2024-03-01T23-15-36_HLE_days3/checkpoints/epoch=000119.ckpt'
|
55 |
+
config_path = '/globalscratch/mridul/ldm/final_runs_eccv/fishes/2024-03-01T23-15-36_HLE_days3/configs/2024-03-01T23-15-36-project.yaml'
|
56 |
+
|
57 |
+
|
58 |
+
|
59 |
+
|
60 |
+
label_to_class_mapping = {0: 'Alosa-chrysochloris', 1: 'Carassius-auratus', 2: 'Cyprinus-carpio', 3: 'Esox-americanus',
|
61 |
+
4: 'Gambusia-affinis', 5: 'Lepisosteus-osseus', 6: 'Lepisosteus-platostomus', 7: 'Lepomis-auritus', 8: 'Lepomis-cyanellus',
|
62 |
+
9: 'Lepomis-gibbosus', 10: 'Lepomis-gulosus', 11: 'Lepomis-humilis', 12: 'Lepomis-macrochirus', 13: 'Lepomis-megalotis',
|
63 |
+
14: 'Lepomis-microlophus', 15: 'Morone-chrysops', 16: 'Morone-mississippiensis', 17: 'Notropis-atherinoides',
|
64 |
+
18: 'Notropis-blennius', 19: 'Notropis-boops', 20: 'Notropis-buccatus', 21: 'Notropis-buchanani', 22: 'Notropis-dorsalis',
|
65 |
+
23: 'Notropis-hudsonius', 24: 'Notropis-leuciodus', 25: 'Notropis-nubilus', 26: 'Notropis-percobromus',
|
66 |
+
27: 'Notropis-stramineus', 28: 'Notropis-telescopus', 29: 'Notropis-texanus', 30: 'Notropis-volucellus',
|
67 |
+
31: 'Notropis-wickliffi', 32: 'Noturus-exilis', 33: 'Noturus-flavus', 34: 'Noturus-gyrinus', 35: 'Noturus-miurus',
|
68 |
+
36: 'Noturus-nocturnus', 37: 'Phenacobius-mirabilis'}
|
69 |
+
|
70 |
+
def get_label_from_class(class_name):
|
71 |
+
for key, value in label_to_class_mapping.items():
|
72 |
+
if value == class_name:
|
73 |
+
return key
|
74 |
+
|
75 |
+
config = OmegaConf.load(config_path) # TODO: Optionally download from same location as ckpt and chnage this logic
|
76 |
+
model = load_model_from_config(config, ckpt_path) # TODO: check path
|
77 |
+
|
78 |
+
device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
|
79 |
+
model = model.to(device)
|
80 |
+
|
81 |
+
if opt.plms:
|
82 |
+
sampler = PLMSSampler(model)
|
83 |
+
else:
|
84 |
+
sampler = DDIMSampler(model)
|
85 |
+
|
86 |
+
os.makedirs(opt.outdir, exist_ok=True)
|
87 |
+
outpath = opt.outdir
|
88 |
+
|
89 |
+
prompt = opt.prompt
|
90 |
+
all_images = []
|
91 |
+
labels = []
|
92 |
+
|
93 |
+
class_to_node = '/fastscratch/mridul/fishes/class_to_ancestral_label.pkl'
|
94 |
+
with open(class_to_node, 'rb') as pickle_file:
|
95 |
+
class_to_node_dict = pickle.load(pickle_file)
|
96 |
+
|
97 |
+
class_to_node_dict = {key.lower(): value for key, value in class_to_node_dict.items()}
|
98 |
+
|
99 |
+
sample_path = os.path.join(outpath, opt.output_dir_name)
|
100 |
+
os.makedirs(sample_path, exist_ok=True)
|
101 |
+
base_count = len(os.listdir(sample_path))
|
102 |
+
|
103 |
+
prompt = class_to_node_dict[fish_name]
|
104 |
+
if swap_fish_name:
|
105 |
+
swap_level = int(swap_level_input.split(" ")[-1]) - 1
|
106 |
+
swap_fish = class_to_node_dict[swap_fish_name]
|
107 |
+
|
108 |
+
swap_fish_split = swap_fish[0].split(',')
|
109 |
+
fish_name_split = prompt[0].split(',')
|
110 |
+
# print(swap_fish_split, fish_name_split)
|
111 |
+
# print(swap_level)
|
112 |
+
fish_name_split[swap_level] = swap_fish_split[swap_level]
|
113 |
+
|
114 |
+
prompt = [','.join(fish_name_split)]
|
115 |
+
|
116 |
+
all_samples=list()
|
117 |
+
with torch.no_grad():
|
118 |
+
with model.ema_scope():
|
119 |
+
uc = None
|
120 |
+
for n in trange(opt.n_iter, desc="Sampling"):
|
121 |
|
122 |
+
all_prompts = opt.n_samples * (prompt)
|
123 |
+
all_prompts = [tuple(all_prompts)]
|
124 |
+
c = model.get_learned_conditioning({'class_to_node': all_prompts})
|
125 |
+
if masking_level_input != "None":
|
126 |
+
masked_level = int(masking_level_input.split(" ")[-1])
|
127 |
+
masked_level = 4-masked_level
|
128 |
+
c = masking_embed(c, levels=masked_level)
|
129 |
+
shape = [3, 64, 64]
|
130 |
+
samples_ddim, _ = sampler.sample(S=opt.ddim_steps,
|
131 |
+
conditioning=c,
|
132 |
+
batch_size=opt.n_samples,
|
133 |
+
shape=shape,
|
134 |
+
verbose=False,
|
135 |
+
unconditional_guidance_scale=opt.scale,
|
136 |
+
unconditional_conditioning=uc,
|
137 |
+
eta=opt.ddim_eta)
|
138 |
+
|
139 |
+
x_samples_ddim = model.decode_first_stage(samples_ddim)
|
140 |
+
x_samples_ddim = torch.clamp((x_samples_ddim+1.0)/2.0, min=0.0, max=1.0)
|
141 |
+
|
142 |
+
all_samples.append(x_samples_ddim)
|
143 |
+
|
144 |
+
###### to make grid
|
145 |
+
# additionally, save as grid
|
146 |
+
grid = torch.stack(all_samples, 0)
|
147 |
+
grid = rearrange(grid, 'n b c h w -> (n b) c h w')
|
148 |
+
grid = make_grid(grid, nrow=opt.n_samples)
|
149 |
+
|
150 |
+
# to image
|
151 |
+
grid = 255. * rearrange(grid, 'c h w -> h w c').cpu().numpy()
|
152 |
+
final_image = Image.fromarray(grid.astype(np.uint8))
|
153 |
+
# final_image.save(os.path.join(sample_path, f'{class_name.replace(" ", "-")}.png'))
|
154 |
+
|
155 |
+
return final_image
|
156 |
|
157 |
|
158 |
if __name__ == "__main__":
|
159 |
+
parser = argparse.ArgumentParser()
|
160 |
+
|
161 |
+
parser.add_argument(
|
162 |
+
"--prompt",
|
163 |
+
type=str,
|
164 |
+
nargs="?",
|
165 |
+
default="a painting of a virus monster playing guitar",
|
166 |
+
help="the prompt to render"
|
167 |
+
)
|
168 |
+
|
169 |
+
parser.add_argument(
|
170 |
+
"--outdir",
|
171 |
+
type=str,
|
172 |
+
nargs="?",
|
173 |
+
help="dir to write results to",
|
174 |
+
default="outputs/txt2img-samples"
|
175 |
+
)
|
176 |
+
parser.add_argument(
|
177 |
+
"--ddim_steps",
|
178 |
+
type=int,
|
179 |
+
default=200,
|
180 |
+
help="number of ddim sampling steps",
|
181 |
+
)
|
182 |
+
|
183 |
+
parser.add_argument(
|
184 |
+
"--plms",
|
185 |
+
action='store_true',
|
186 |
+
help="use plms sampling",
|
187 |
+
)
|
188 |
+
|
189 |
+
parser.add_argument(
|
190 |
+
"--ddim_eta",
|
191 |
+
type=float,
|
192 |
+
default=1.0,
|
193 |
+
help="ddim eta (eta=0.0 corresponds to deterministic sampling",
|
194 |
+
)
|
195 |
+
parser.add_argument(
|
196 |
+
"--n_iter",
|
197 |
+
type=int,
|
198 |
+
default=1,
|
199 |
+
help="sample this often",
|
200 |
+
)
|
201 |
+
|
202 |
+
parser.add_argument(
|
203 |
+
"--H",
|
204 |
+
type=int,
|
205 |
+
default=256,
|
206 |
+
help="image height, in pixel space",
|
207 |
+
)
|
208 |
+
|
209 |
+
parser.add_argument(
|
210 |
+
"--W",
|
211 |
+
type=int,
|
212 |
+
default=256,
|
213 |
+
help="image width, in pixel space",
|
214 |
+
)
|
215 |
+
|
216 |
+
parser.add_argument(
|
217 |
+
"--n_samples",
|
218 |
+
type=int,
|
219 |
+
default=1,
|
220 |
+
help="how many samples to produce for the given prompt",
|
221 |
+
)
|
222 |
+
|
223 |
+
parser.add_argument(
|
224 |
+
"--output_dir_name",
|
225 |
+
type=str,
|
226 |
+
default='default_file',
|
227 |
+
help="name of folder",
|
228 |
+
)
|
229 |
+
|
230 |
+
parser.add_argument(
|
231 |
+
"--postfix",
|
232 |
+
type=str,
|
233 |
+
default='',
|
234 |
+
help="name of folder",
|
235 |
+
)
|
236 |
+
|
237 |
+
parser.add_argument(
|
238 |
+
"--scale",
|
239 |
+
type=float,
|
240 |
+
# default=5.0,
|
241 |
+
default=1.0,
|
242 |
+
help="unconditional guidance scale: eps = eps(x, empty) + scale * (eps(x, cond) - eps(x, empty))",
|
243 |
+
)
|
244 |
+
opt = parser.parse_args()
|
245 |
+
|
246 |
+
|
247 |
+
def setup_interface():
|
248 |
+
with gr.Blocks() as demo:
|
249 |
+
with gr.Row():
|
250 |
+
with gr.Column():
|
251 |
+
gr.Markdown("### Generate Images Based on Prompts")
|
252 |
+
gr.Markdown("Enter a prompt to generate an image:")
|
253 |
+
prompt_input = gr.Textbox(label="Species Name")
|
254 |
+
gr.Markdown("Trait Masking")
|
255 |
+
with gr.Row():
|
256 |
+
masking_level_input = gr.Dropdown(label="Select Ancestral Level", choices=["None", "Level 3", "Level 2"], value="None")
|
257 |
+
# masking_node_input = gr.Dropdown(label="Select Internal", choices=["0", "1", "2", "3", "4", "5", "6", "7", "8"], value="0")
|
258 |
+
|
259 |
+
gr.Markdown("Trait Swapping")
|
260 |
+
with gr.Row():
|
261 |
+
swap_fish_name = gr.Textbox(label="Species Name to swap trait with:")
|
262 |
+
swap_level_input = gr.Dropdown(label="Level of swapping", choices=["Level 3", "Level 2"], value="Level 3")
|
263 |
+
submit_button = gr.Button("Generate")
|
264 |
+
gr.Markdown("### Phylogeny Tree")
|
265 |
+
architecture_image = "phylogeny_tree.jpg" # Update this with the actual path
|
266 |
+
gr.Image(value=architecture_image, label="Phylogeny Tree")
|
267 |
+
|
268 |
+
with gr.Column():
|
269 |
+
gr.Markdown("### Generated Image")
|
270 |
+
output_image = gr.Image(label="Generated Image")
|
271 |
+
|
272 |
+
# Display an image of the architecture
|
273 |
+
|
274 |
+
|
275 |
+
submit_button.click(
|
276 |
+
fn=generate_image,
|
277 |
+
inputs=[prompt_input, masking_level_input,
|
278 |
+
swap_fish_name, swap_level_input],
|
279 |
+
outputs=output_image
|
280 |
+
)
|
281 |
+
|
282 |
+
return demo
|
283 |
+
|
284 |
+
# # Launch the interface
|
285 |
+
# iface = setup_interface()
|
286 |
+
|
287 |
+
# iface = gr.Interface(
|
288 |
+
# fn=generate_image,
|
289 |
+
# inputs=gr.Textbox(label="Prompt"),
|
290 |
+
# outputs=[
|
291 |
+
# gr.Image(label="Generated Image"),
|
292 |
+
# ]
|
293 |
+
# )
|
294 |
+
|
295 |
+
iface = setup_interface()
|
296 |
+
|
297 |
+
iface.launch(share=True)
|
ldm/models/__pycache__/autoencoder.cpython-38.pyc
CHANGED
Binary files a/ldm/models/__pycache__/autoencoder.cpython-38.pyc and b/ldm/models/__pycache__/autoencoder.cpython-38.pyc differ
|
|
ldm/models/diffusion/__pycache__/__init__.cpython-38.pyc
CHANGED
Binary files a/ldm/models/diffusion/__pycache__/__init__.cpython-38.pyc and b/ldm/models/diffusion/__pycache__/__init__.cpython-38.pyc differ
|
|
ldm/models/diffusion/__pycache__/ddim.cpython-38.pyc
CHANGED
Binary files a/ldm/models/diffusion/__pycache__/ddim.cpython-38.pyc and b/ldm/models/diffusion/__pycache__/ddim.cpython-38.pyc differ
|
|
ldm/models/diffusion/__pycache__/ddpm.cpython-38.pyc
CHANGED
Binary files a/ldm/models/diffusion/__pycache__/ddpm.cpython-38.pyc and b/ldm/models/diffusion/__pycache__/ddpm.cpython-38.pyc differ
|
|
ldm/models/diffusion/__pycache__/plms.cpython-38.pyc
CHANGED
Binary files a/ldm/models/diffusion/__pycache__/plms.cpython-38.pyc and b/ldm/models/diffusion/__pycache__/plms.cpython-38.pyc differ
|
|
ldm/models/disentanglement/__pycache__/iterative_normalization.cpython-38.pyc
CHANGED
Binary files a/ldm/models/disentanglement/__pycache__/iterative_normalization.cpython-38.pyc and b/ldm/models/disentanglement/__pycache__/iterative_normalization.cpython-38.pyc differ
|
|
ldm/modules/diffusionmodules/__pycache__/__init__.cpython-38.pyc
CHANGED
Binary files a/ldm/modules/diffusionmodules/__pycache__/__init__.cpython-38.pyc and b/ldm/modules/diffusionmodules/__pycache__/__init__.cpython-38.pyc differ
|
|
ldm/modules/diffusionmodules/__pycache__/model.cpython-38.pyc
CHANGED
Binary files a/ldm/modules/diffusionmodules/__pycache__/model.cpython-38.pyc and b/ldm/modules/diffusionmodules/__pycache__/model.cpython-38.pyc differ
|
|
ldm/modules/diffusionmodules/__pycache__/openaimodel.cpython-38.pyc
CHANGED
Binary files a/ldm/modules/diffusionmodules/__pycache__/openaimodel.cpython-38.pyc and b/ldm/modules/diffusionmodules/__pycache__/openaimodel.cpython-38.pyc differ
|
|
ldm/modules/diffusionmodules/__pycache__/util.cpython-38.pyc
CHANGED
Binary files a/ldm/modules/diffusionmodules/__pycache__/util.cpython-38.pyc and b/ldm/modules/diffusionmodules/__pycache__/util.cpython-38.pyc differ
|
|
ldm/modules/distributions/__pycache__/__init__.cpython-38.pyc
CHANGED
Binary files a/ldm/modules/distributions/__pycache__/__init__.cpython-38.pyc and b/ldm/modules/distributions/__pycache__/__init__.cpython-38.pyc differ
|
|
ldm/modules/distributions/__pycache__/distributions.cpython-38.pyc
CHANGED
Binary files a/ldm/modules/distributions/__pycache__/distributions.cpython-38.pyc and b/ldm/modules/distributions/__pycache__/distributions.cpython-38.pyc differ
|
|
ldm/modules/encoders/__pycache__/__init__.cpython-38.pyc
CHANGED
Binary files a/ldm/modules/encoders/__pycache__/__init__.cpython-38.pyc and b/ldm/modules/encoders/__pycache__/__init__.cpython-38.pyc differ
|
|
ldm/modules/encoders/__pycache__/modules.cpython-38.pyc
CHANGED
Binary files a/ldm/modules/encoders/__pycache__/modules.cpython-38.pyc and b/ldm/modules/encoders/__pycache__/modules.cpython-38.pyc differ
|
|
phylogeny_tree.jpg
ADDED
phylogeny_tree.pdf
ADDED
Binary file (236 kB). View file
|
|
phylogeny_tree.png
ADDED
sample_level_encoding.py
CHANGED
@@ -32,6 +32,16 @@ def load_model_from_config(config, ckpt, verbose=False):
|
|
32 |
model.eval()
|
33 |
return model
|
34 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
35 |
|
36 |
if __name__ == "__main__":
|
37 |
parser = argparse.ArgumentParser()
|
@@ -206,6 +216,9 @@ if __name__ == "__main__":
|
|
206 |
with open(class_to_node, 'rb') as pickle_file:
|
207 |
class_to_node_dict = pickle.load(pickle_file)
|
208 |
|
|
|
|
|
|
|
209 |
sample_path = os.path.join(outpath, opt.output_dir_name)
|
210 |
os.makedirs(sample_path, exist_ok=True)
|
211 |
base_count = len(os.listdir(sample_path))
|
@@ -223,6 +236,7 @@ if __name__ == "__main__":
|
|
223 |
all_prompts = opt.n_samples * (prompt)
|
224 |
all_prompts = [tuple(all_prompts)]
|
225 |
print(class_name, prompt)
|
|
|
226 |
c = model.get_learned_conditioning({'class_to_node': all_prompts})
|
227 |
shape = [3, 64, 64]
|
228 |
samples_ddim, _ = sampler.sample(S=opt.ddim_steps,
|
|
|
32 |
model.eval()
|
33 |
return model
|
34 |
|
35 |
+
def masking_embed(embedding, levels=1):
|
36 |
+
"""
|
37 |
+
size of embedding - nx1xd, n: number of samples, d - 512
|
38 |
+
replacing the last 128*levels from the embedding
|
39 |
+
"""
|
40 |
+
replace_size = 128*levels
|
41 |
+
random_noise = torch.randn(embedding.shape[0], embedding.shape[1], replace_size)
|
42 |
+
embedding[:, :, -replace_size:] = random_noise
|
43 |
+
return embedding
|
44 |
+
|
45 |
|
46 |
if __name__ == "__main__":
|
47 |
parser = argparse.ArgumentParser()
|
|
|
216 |
with open(class_to_node, 'rb') as pickle_file:
|
217 |
class_to_node_dict = pickle.load(pickle_file)
|
218 |
|
219 |
+
class_to_node_dict = {key.lower(): value for key, value in class_to_node_dict.items()}
|
220 |
+
|
221 |
+
|
222 |
sample_path = os.path.join(outpath, opt.output_dir_name)
|
223 |
os.makedirs(sample_path, exist_ok=True)
|
224 |
base_count = len(os.listdir(sample_path))
|
|
|
236 |
all_prompts = opt.n_samples * (prompt)
|
237 |
all_prompts = [tuple(all_prompts)]
|
238 |
print(class_name, prompt)
|
239 |
+
breakpoint()
|
240 |
c = model.get_learned_conditioning({'class_to_node': all_prompts})
|
241 |
shape = [3, 64, 64]
|
242 |
samples_ddim, _ = sampler.sample(S=opt.ddim_steps,
|