Spaces:
Running
on
Zero
Running
on
Zero
Update app.py
Browse files
app.py
CHANGED
@@ -36,23 +36,10 @@ sdxl_pipe = DiffusionPipeline.from_pretrained(
|
|
36 |
)
|
37 |
sdxl_pipe.to(device)
|
38 |
|
39 |
-
# superprompt-v1
|
40 |
-
tokenizer = T5Tokenizer.from_pretrained("roborovski/superprompt-v1")
|
41 |
-
model = T5ForConditionalGeneration.from_pretrained(
|
42 |
-
"roborovski/superprompt-v1", device_map="auto", torch_dtype="auto"
|
43 |
-
)
|
44 |
-
model.to(device)
|
45 |
-
|
46 |
-
# toggle visibility the enhanced prompt output
|
47 |
-
def update_visibility(enhance_prompt):
|
48 |
-
return gr.update(visible=enhance_prompt)
|
49 |
-
|
50 |
-
|
51 |
# Define the image generation function for the Arena tab
|
52 |
@spaces.GPU(duration=80)
|
53 |
def generate_arena_images(
|
54 |
prompt,
|
55 |
-
enhance_prompt,
|
56 |
negative_prompt,
|
57 |
num_inference_steps,
|
58 |
height,
|
@@ -67,23 +54,6 @@ def generate_arena_images(
|
|
67 |
if seed == 0:
|
68 |
seed = random.randint(1, 2**32 - 1)
|
69 |
|
70 |
-
if enhance_prompt:
|
71 |
-
transformers.set_seed(seed)
|
72 |
-
|
73 |
-
input_text = f"Expand the following prompt to add more detail: {prompt}"
|
74 |
-
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to(device)
|
75 |
-
|
76 |
-
outputs = model.generate(
|
77 |
-
input_ids,
|
78 |
-
max_new_tokens=512,
|
79 |
-
repetition_penalty=1.2,
|
80 |
-
do_sample=True,
|
81 |
-
temperature=0.7,
|
82 |
-
top_p=1,
|
83 |
-
top_k=50,
|
84 |
-
)
|
85 |
-
prompt = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
86 |
-
|
87 |
generator = torch.Generator().manual_seed(seed)
|
88 |
|
89 |
# Generate images for both models
|
@@ -112,7 +82,7 @@ def generate_arena_images(
|
|
112 |
generator,
|
113 |
)
|
114 |
|
115 |
-
return images_1, images_2
|
116 |
|
117 |
|
118 |
# Helper function to generate images for a single model
|
@@ -155,7 +125,6 @@ def generate_single_image(
|
|
155 |
@spaces.GPU(duration=80)
|
156 |
def generate_individual_image(
|
157 |
prompt,
|
158 |
-
enhance_prompt,
|
159 |
negative_prompt,
|
160 |
num_inference_steps,
|
161 |
height,
|
@@ -169,23 +138,6 @@ def generate_individual_image(
|
|
169 |
if seed == 0:
|
170 |
seed = random.randint(1, 2**32 - 1)
|
171 |
|
172 |
-
if enhance_prompt:
|
173 |
-
transformers.set_seed(seed)
|
174 |
-
|
175 |
-
input_text = f"Expand the following prompt to add more detail: {prompt}"
|
176 |
-
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to(device)
|
177 |
-
|
178 |
-
outputs = model.generate(
|
179 |
-
input_ids,
|
180 |
-
max_new_tokens=512,
|
181 |
-
repetition_penalty=1.2,
|
182 |
-
do_sample=True,
|
183 |
-
temperature=0.7,
|
184 |
-
top_p=1,
|
185 |
-
top_k=50,
|
186 |
-
)
|
187 |
-
prompt = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
188 |
-
|
189 |
generator = torch.Generator().manual_seed(seed)
|
190 |
|
191 |
output = generate_single_image(
|
@@ -201,14 +153,14 @@ def generate_individual_image(
|
|
201 |
generator,
|
202 |
)
|
203 |
|
204 |
-
return output
|
205 |
|
206 |
|
207 |
# Create the Gradio interface
|
208 |
examples = [
|
209 |
-
["A white car racing fast to the moon."
|
210 |
-
["A woman in a red dress singing on top of a building."
|
211 |
-
["An astronaut on mars in a futuristic cyborg suit."
|
212 |
]
|
213 |
|
214 |
css = """
|
@@ -240,9 +192,6 @@ with gr.Blocks(css=css) as demo:
|
|
240 |
info="Describe the image you want",
|
241 |
placeholder="A cat...",
|
242 |
)
|
243 |
-
enhance_prompt = gr.Checkbox(
|
244 |
-
label="Prompt Enhancement with SuperPrompt-v1", value=True
|
245 |
-
)
|
246 |
model_choice_1 = gr.Dropdown(
|
247 |
label="Stable Diffusion Model 1",
|
248 |
choices=["sd3 medium", "sd2.1", "sdxl"],
|
@@ -256,14 +205,6 @@ with gr.Blocks(css=css) as demo:
|
|
256 |
run_button = gr.Button("Run")
|
257 |
result_1 = gr.Gallery(label="Generated Images (Model 1)", elem_id="gallery_1")
|
258 |
result_2 = gr.Gallery(label="Generated Images (Model 2)", elem_id="gallery_2")
|
259 |
-
better_prompt = gr.Textbox(
|
260 |
-
label="Enhanced Prompt",
|
261 |
-
info="The output of your enhanced prompt used for the Image Generation",
|
262 |
-
visible=True,
|
263 |
-
)
|
264 |
-
enhance_prompt.change(
|
265 |
-
fn=update_visibility, inputs=enhance_prompt, outputs=better_prompt
|
266 |
-
)
|
267 |
with gr.Accordion("Advanced options", open=False):
|
268 |
with gr.Row():
|
269 |
negative_prompt = gr.Textbox(
|
@@ -326,8 +267,8 @@ with gr.Blocks(css=css) as demo:
|
|
326 |
|
327 |
gr.Examples(
|
328 |
examples=examples,
|
329 |
-
inputs=[prompt
|
330 |
-
outputs=[result_1, result_2
|
331 |
fn=generate_arena_images,
|
332 |
)
|
333 |
|
@@ -339,7 +280,6 @@ with gr.Blocks(css=css) as demo:
|
|
339 |
fn=generate_arena_images,
|
340 |
inputs=[
|
341 |
prompt,
|
342 |
-
enhance_prompt,
|
343 |
negative_prompt,
|
344 |
num_inference_steps,
|
345 |
width,
|
@@ -350,7 +290,7 @@ with gr.Blocks(css=css) as demo:
|
|
350 |
model_choice_1,
|
351 |
model_choice_2,
|
352 |
],
|
353 |
-
outputs=[result_1, result_2
|
354 |
)
|
355 |
|
356 |
with gr.TabItem("Individual"):
|
@@ -361,9 +301,6 @@ with gr.Blocks(css=css) as demo:
|
|
361 |
info="Describe the image you want",
|
362 |
placeholder="A cat...",
|
363 |
)
|
364 |
-
enhance_prompt = gr.Checkbox(
|
365 |
-
label="Prompt Enhancement with SuperPrompt-v1", value=True
|
366 |
-
)
|
367 |
model_choice = gr.Dropdown(
|
368 |
label="Stable Diffusion Model",
|
369 |
choices=["sd3 medium", "sd2.1", "sdxl"],
|
@@ -371,14 +308,6 @@ with gr.Blocks(css=css) as demo:
|
|
371 |
)
|
372 |
run_button = gr.Button("Run")
|
373 |
result = gr.Gallery(label="Generated AI Images", elem_id="gallery")
|
374 |
-
better_prompt = gr.Textbox(
|
375 |
-
label="Enhanced Prompt",
|
376 |
-
info="The output of your enhanced prompt used for the Image Generation",
|
377 |
-
visible=True,
|
378 |
-
)
|
379 |
-
enhance_prompt.change(
|
380 |
-
fn=update_visibility, inputs=enhance_prompt, outputs=better_prompt
|
381 |
-
)
|
382 |
with gr.Accordion("Advanced options", open=False):
|
383 |
with gr.Row():
|
384 |
negative_prompt = gr.Textbox(
|
@@ -441,8 +370,8 @@ with gr.Blocks(css=css) as demo:
|
|
441 |
|
442 |
gr.Examples(
|
443 |
examples=examples,
|
444 |
-
inputs=[prompt
|
445 |
-
outputs=[result
|
446 |
fn=generate_individual_image,
|
447 |
)
|
448 |
|
@@ -454,7 +383,6 @@ with gr.Blocks(css=css) as demo:
|
|
454 |
fn=generate_individual_image,
|
455 |
inputs=[
|
456 |
prompt,
|
457 |
-
enhance_prompt,
|
458 |
negative_prompt,
|
459 |
num_inference_steps,
|
460 |
width,
|
@@ -464,7 +392,7 @@ with gr.Blocks(css=css) as demo:
|
|
464 |
num_images_per_prompt,
|
465 |
model_choice,
|
466 |
],
|
467 |
-
outputs=[result
|
468 |
)
|
469 |
|
470 |
demo.queue().launch(share=False)
|
|
|
36 |
)
|
37 |
sdxl_pipe.to(device)
|
38 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
39 |
# Define the image generation function for the Arena tab
|
40 |
@spaces.GPU(duration=80)
|
41 |
def generate_arena_images(
|
42 |
prompt,
|
|
|
43 |
negative_prompt,
|
44 |
num_inference_steps,
|
45 |
height,
|
|
|
54 |
if seed == 0:
|
55 |
seed = random.randint(1, 2**32 - 1)
|
56 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
57 |
generator = torch.Generator().manual_seed(seed)
|
58 |
|
59 |
# Generate images for both models
|
|
|
82 |
generator,
|
83 |
)
|
84 |
|
85 |
+
return images_1, images_2
|
86 |
|
87 |
|
88 |
# Helper function to generate images for a single model
|
|
|
125 |
@spaces.GPU(duration=80)
|
126 |
def generate_individual_image(
|
127 |
prompt,
|
|
|
128 |
negative_prompt,
|
129 |
num_inference_steps,
|
130 |
height,
|
|
|
138 |
if seed == 0:
|
139 |
seed = random.randint(1, 2**32 - 1)
|
140 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
141 |
generator = torch.Generator().manual_seed(seed)
|
142 |
|
143 |
output = generate_single_image(
|
|
|
153 |
generator,
|
154 |
)
|
155 |
|
156 |
+
return output
|
157 |
|
158 |
|
159 |
# Create the Gradio interface
|
160 |
examples = [
|
161 |
+
["A white car racing fast to the moon."],
|
162 |
+
["A woman in a red dress singing on top of a building."],
|
163 |
+
["An astronaut on mars in a futuristic cyborg suit."],
|
164 |
]
|
165 |
|
166 |
css = """
|
|
|
192 |
info="Describe the image you want",
|
193 |
placeholder="A cat...",
|
194 |
)
|
|
|
|
|
|
|
195 |
model_choice_1 = gr.Dropdown(
|
196 |
label="Stable Diffusion Model 1",
|
197 |
choices=["sd3 medium", "sd2.1", "sdxl"],
|
|
|
205 |
run_button = gr.Button("Run")
|
206 |
result_1 = gr.Gallery(label="Generated Images (Model 1)", elem_id="gallery_1")
|
207 |
result_2 = gr.Gallery(label="Generated Images (Model 2)", elem_id="gallery_2")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
208 |
with gr.Accordion("Advanced options", open=False):
|
209 |
with gr.Row():
|
210 |
negative_prompt = gr.Textbox(
|
|
|
267 |
|
268 |
gr.Examples(
|
269 |
examples=examples,
|
270 |
+
inputs=[prompt],
|
271 |
+
outputs=[result_1, result_2],
|
272 |
fn=generate_arena_images,
|
273 |
)
|
274 |
|
|
|
280 |
fn=generate_arena_images,
|
281 |
inputs=[
|
282 |
prompt,
|
|
|
283 |
negative_prompt,
|
284 |
num_inference_steps,
|
285 |
width,
|
|
|
290 |
model_choice_1,
|
291 |
model_choice_2,
|
292 |
],
|
293 |
+
outputs=[result_1, result_2],
|
294 |
)
|
295 |
|
296 |
with gr.TabItem("Individual"):
|
|
|
301 |
info="Describe the image you want",
|
302 |
placeholder="A cat...",
|
303 |
)
|
|
|
|
|
|
|
304 |
model_choice = gr.Dropdown(
|
305 |
label="Stable Diffusion Model",
|
306 |
choices=["sd3 medium", "sd2.1", "sdxl"],
|
|
|
308 |
)
|
309 |
run_button = gr.Button("Run")
|
310 |
result = gr.Gallery(label="Generated AI Images", elem_id="gallery")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
311 |
with gr.Accordion("Advanced options", open=False):
|
312 |
with gr.Row():
|
313 |
negative_prompt = gr.Textbox(
|
|
|
370 |
|
371 |
gr.Examples(
|
372 |
examples=examples,
|
373 |
+
inputs=[prompt],
|
374 |
+
outputs=[result],
|
375 |
fn=generate_individual_image,
|
376 |
)
|
377 |
|
|
|
383 |
fn=generate_individual_image,
|
384 |
inputs=[
|
385 |
prompt,
|
|
|
386 |
negative_prompt,
|
387 |
num_inference_steps,
|
388 |
width,
|
|
|
392 |
num_images_per_prompt,
|
393 |
model_choice,
|
394 |
],
|
395 |
+
outputs=[result],
|
396 |
)
|
397 |
|
398 |
demo.queue().launch(share=False)
|