Upload 4 files

README.md

@@ -1,5 +1,5 @@
 ---
-title: FLUX.1 [dev]
+title: FLUX.1 [dev] sigmas test
 emoji: 🖥️
 colorFrom: yellow
 colorTo: pink

app.py

@@ -7,13 +7,14 @@ from diffusers import  DiffusionPipeline, FlowMatchEulerDiscreteScheduler, Autoe
 from transformers import CLIPTextModel, CLIPTokenizer,T5EncoderModel, T5TokenizerFast
 from live_preview_helpers import calculate_shift, retrieve_timesteps, flux_pipe_call_that_returns_an_iterable_of_images
 from gradio_imageslider import ImageSlider
+from PIL import Image, ImageDraw
 
 dtype = torch.bfloat16
 #model_id = "black-forest-labs/FLUX.1-dev"
 model_id = "camenduru/FLUX.1-dev-diffusers"
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
-taef1 = AutoencoderTiny.from_pretrained("madebyollin/taef1", torch_dtype=dtype).to(device)
+#taef1 = AutoencoderTiny.from_pretrained("madebyollin/taef1", torch_dtype=dtype).to(device)
 good_vae = AutoencoderKL.from_pretrained(model_id, subfolder="vae", torch_dtype=dtype).to(device)
 #pipe = DiffusionPipeline.from_pretrained(model_id, torch_dtype=dtype, vae=taef1).to(device)
 pipe = DiffusionPipeline.from_pretrained(model_id, torch_dtype=dtype, vae=good_vae).to(device)
@@ -24,23 +25,24 @@ MAX_IMAGE_SIZE = 2048
 
 pipe.flux_pipe_call_that_returns_an_iterable_of_images = flux_pipe_call_that_returns_an_iterable_of_images.__get__(pipe)
 
+def get_cmp_image(im1: Image.Image, im2: Image.Image, sigmas: float):
+    dst = Image.new('RGB', (im1.width + im2.width, im1.height))
+    dst.paste(im1.convert('RGB'), (0, 0))
+    dst.paste(im2.convert('RGB'), (im1.width, 0))
+    draw = ImageDraw.Draw(dst)
+    draw.text((64, 64), 'Sigmas: 1.0', 'red')
+    draw.text((im1.width + 64, 64), f'Sigmas: {sigmas}', 'red')
+    return dst
+
 @spaces.GPU(duration=90)
-def infer(prompt, seed=42, randomize_seed=False, width=1024, height=1024, guidance_scale=3.5, num_inference_steps=28, sigmas=0.95, progress=gr.Progress(track_tqdm=True)):
+def infer(prompt, seed=42, randomize_seed=False, width=1024, height=1024, guidance_scale=3.5, num_inference_steps=28, mul_sigmas=0.95, is_cmp=True, progress=gr.Progress(track_tqdm=True)):
     if randomize_seed:
         seed = random.randint(0, MAX_SEED)
     generator = torch.Generator().manual_seed(seed)
     
-    #for img in pipe.flux_pipe_call_that_returns_an_iterable_of_images(
-    image_def = pipe(
-            prompt=prompt,
-            guidance_scale=guidance_scale,
-            num_inference_steps=num_inference_steps,
-            width=width,
-            height=height,
-            generator=generator,
-            output_type="pil",
-        ).images[0]
-    #        yield img, seed
+    sigmas = np.linspace(1.0, 1 / num_inference_steps, num_inference_steps)
+    sigmas = sigmas * mul_sigmas
+
     image_sigmas = pipe(
             prompt=prompt,
             guidance_scale=guidance_scale,
@@ -49,10 +51,22 @@ def infer(prompt, seed=42, randomize_seed=False, width=1024, height=1024, guidan
             height=height,
             generator=generator,
             output_type="pil",
-            mul_sigmas=sigmas
+            sigmas=sigmas
         ).images[0]
-    return [image_def, image_sigmas], seed
     
+    if is_cmp:
+        image_def = pipe(
+                prompt=prompt,
+                guidance_scale=guidance_scale,
+                num_inference_steps=num_inference_steps,
+                width=width,
+                height=height,
+                generator=generator,
+                output_type="pil",
+            ).images[0]
+        return [image_def, image_sigmas], get_cmp_image(image_def, image_sigmas), seed
+    else: return [image_sigmas, image_sigmas], None, seed
+
 examples = [
     "a tiny astronaut hatching from an egg on the moon",
     "a cat holding a sign that says hello world",
@@ -87,16 +101,19 @@ with gr.Blocks(css=css) as demo:
             run_button = gr.Button("Run", scale=0)
         
         #result = gr.Image(label="Result", show_label=False)
-        result = ImageSlider(label="Result", show_label=False, type="pil", slider_color="pink")
+        result = ImageSlider(label="Result", show_label=False, type="pil", slider_color="pink", format="png")
+        result_cmp = gr.Image(label="Result (comparing)", show_label=False, type="pil", format="png", height=256)
         
         with gr.Accordion("Advanced Settings", open=True):
-            sigmas = gr.Slider(
-                label="Sigmas",
-                minimum=0,
-                maximum=1.0,
-                step=0.01,
-                value=0.95,
-            )
+            with gr.Row():
+                sigmas = gr.Slider(
+                    label="Sigmas",
+                    minimum=0,
+                    maximum=1.0,
+                    step=0.01,
+                    value=0.95,
+                )
+                is_cmp = gr.Checkbox("Compare images with/without sigmas")
             
             seed = gr.Slider(
                 label="Seed",
@@ -148,15 +165,15 @@ with gr.Blocks(css=css) as demo:
             examples = examples,
             fn = infer,
             inputs = [prompt],
-            outputs = [result, seed],
+            outputs = [result, result_cmp, seed],
             cache_examples="lazy"
         )
 
     gr.on(
         triggers=[run_button.click, prompt.submit],
         fn = infer,
-        inputs = [prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps, sigmas],
-        outputs = [result, seed]
+        inputs = [prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps, sigmas, is_cmp],
+        outputs = [result, result_cmp, seed]
     )
 
 demo.launch()

requirements.txt

@@ -1,6 +1,6 @@
 accelerate
 #git+https://github.com/huggingface/diffusers.git
-git+https://github.com/huggingface/diffusers.git@6b1c4a766b7f83fe06ddb9bbb58c1122072efefe
+#git+https://github.com/huggingface/diffusers.git@6b1c4a766b7f83fe06ddb9bbb58c1122072efefegit+https://github.com/huggingface/diffusers.git@ad3344e2be033887d854d2731757db8b80dcfb06
 torch
 transformers==4.42.4
 xformers