File size: 10,233 Bytes
394d08e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a659304
 
 
 
 
 
 
 
 
 
 
 
 
394d08e
 
 
 
 
 
 
 
 
 
 
 
592470d
394d08e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
from diffusers import (
    StableDiffusionXLControlNetImg2ImgPipeline,
    ControlNetModel,
    AutoencoderKL,
    AutoencoderTiny,
    LCMScheduler,
)
from compel import Compel, ReturnedEmbeddingsType
import torch
from pipelines.utils.canny_gpu import SobelOperator

try:
    import intel_extension_for_pytorch as ipex  # type: ignore
except:
    pass

import psutil
from config import Args
from pydantic import BaseModel, Field
from PIL import Image
import math

controlnet_model = "diffusers/controlnet-canny-sdxl-1.0"
base_model = "segmind/Segmind-Vega"
lora_model = "segmind/Segmind-VegaRT"
taesd_model = "madebyollin/taesdxl"

default_prompt = "Portrait of The Terminator with , glare pose, detailed, intricate, full of colour, cinematic lighting, trending on artstation, 8k, hyperrealistic, focused, extreme details, unreal engine 5 cinematic, masterpiece"
default_negative_prompt = "blurry, low quality, render, 3D, oversaturated"
page_content = """
<h1 class="text-3xl font-bold">Real-Time SegmindVegaRT</h1>
<h3 class="text-xl font-bold">Image-to-Image ControlNet</h3>
<p class="text-sm">
    This demo showcases
    <a
    href="https://huggingface.co/segmind/Segmind-VegaRT"
    target="_blank"
    class="text-blue-500 underline hover:no-underline">Segmind-VegaRT</a>
Image to Image pipeline using
    <a
    href="https://huggingface.co/docs/diffusers/main/en/using-diffusers/sdxl_turbo"
    target="_blank"
    class="text-blue-500 underline hover:no-underline">Diffusers</a
    > with a MJPEG stream server.
</p>
<p class="text-sm text-gray-500">
    Change the prompt to generate different images, accepts <a
    href="https://github.com/damian0815/compel/blob/main/doc/syntax.md"
    target="_blank"
    class="text-blue-500 underline hover:no-underline">Compel</a
    > syntax.
</p>
"""


class Pipeline:
    class Info(BaseModel):
        name: str = "controlnet+SegmindVegaRT"
        title: str = "SegmindVegaRT + Controlnet"
        description: str = "Generates an image from a text prompt"
        input_mode: str = "image"
        page_content: str = page_content

    class InputParams(BaseModel):
        prompt: str = Field(
            default_prompt,
            title="Prompt",
            field="textarea",
            id="prompt",
        )
        negative_prompt: str = Field(
            default_negative_prompt,
            title="Negative Prompt",
            field="textarea",
            id="negative_prompt",
            hide=True,
        )
        seed: int = Field(
            2159232, min=0, title="Seed", field="seed", hide=True, id="seed"
        )
        steps: int = Field(
            2, min=1, max=15, title="Steps", field="range", hide=True, id="steps"
        )
        width: int = Field(
            1024, min=2, max=15, title="Width", disabled=True, hide=True, id="width"
        )
        height: int = Field(
            1024, min=2, max=15, title="Height", disabled=True, hide=True, id="height"
        )
        guidance_scale: float = Field(
            0.0,
            min=0,
            max=1,
            step=0.001,
            title="Guidance Scale",
            field="range",
            hide=True,
            id="guidance_scale",
        )
        strength: float = Field(
            0.5,
            min=0.25,
            max=1.0,
            step=0.001,
            title="Strength",
            field="range",
            hide=True,
            id="strength",
        )
        controlnet_scale: float = Field(
            0.5,
            min=0,
            max=1.0,
            step=0.001,
            title="Controlnet Scale",
            field="range",
            hide=True,
            id="controlnet_scale",
        )
        controlnet_start: float = Field(
            0.0,
            min=0,
            max=1.0,
            step=0.001,
            title="Controlnet Start",
            field="range",
            hide=True,
            id="controlnet_start",
        )
        controlnet_end: float = Field(
            1.0,
            min=0,
            max=1.0,
            step=0.001,
            title="Controlnet End",
            field="range",
            hide=True,
            id="controlnet_end",
        )
        canny_low_threshold: float = Field(
            0.31,
            min=0,
            max=1.0,
            step=0.001,
            title="Canny Low Threshold",
            field="range",
            hide=True,
            id="canny_low_threshold",
        )
        canny_high_threshold: float = Field(
            0.125,
            min=0,
            max=1.0,
            step=0.001,
            title="Canny High Threshold",
            field="range",
            hide=True,
            id="canny_high_threshold",
        )
        debug_canny: bool = Field(
            False,
            title="Debug Canny",
            field="checkbox",
            hide=True,
            id="debug_canny",
        )

    def __init__(self, args: Args, device: torch.device, torch_dtype: torch.dtype):
        controlnet_canny = ControlNetModel.from_pretrained(
            controlnet_model,
            torch_dtype=torch_dtype,
        ).to(device)
        vae = AutoencoderKL.from_pretrained(
            "madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch_dtype
        )
        if args.safety_checker:
            self.pipe = StableDiffusionXLControlNetImg2ImgPipeline.from_pretrained(
                base_model,
                controlnet=controlnet_canny,
                vae=vae,
            )
        else:
            self.pipe = StableDiffusionXLControlNetImg2ImgPipeline.from_pretrained(
                base_model,
                safety_checker=None,
                controlnet=controlnet_canny,
                vae=vae,
            )
        self.canny_torch = SobelOperator(device=device)

        self.pipe.load_lora_weights(lora_model)
        self.pipe.fuse_lora()
        self.pipe.scheduler = LCMScheduler.from_pretrained(
            base_model, subfolder="scheduler"
        )

        if args.sfast:
            from sfast.compilers.stable_diffusion_pipeline_compiler import (
                compile,
                CompilationConfig,
            )

            config = CompilationConfig.Default()
            config.enable_xformers = True
            config.enable_triton = True
            config.enable_cuda_graph = True
            self.pipe = compile(self.pipe, config=config)

        self.pipe.set_progress_bar_config(disable=True)
        self.pipe.to(device=device, dtype=torch_dtype).to(device)
        if device.type != "mps":
            self.pipe.unet.to(memory_format=torch.channels_last)

        if args.compel:
            self.pipe.compel_proc = Compel(
                tokenizer=[self.pipe.tokenizer, self.pipe.tokenizer_2],
                text_encoder=[self.pipe.text_encoder, self.pipe.text_encoder_2],
                returned_embeddings_type=ReturnedEmbeddingsType.PENULTIMATE_HIDDEN_STATES_NON_NORMALIZED,
                requires_pooled=[False, True],
            )
        if args.taesd:
            self.pipe.vae = AutoencoderTiny.from_pretrained(
                taesd_model, torch_dtype=torch_dtype, use_safetensors=True
            ).to(device)

        if args.torch_compile:
            self.pipe.unet = torch.compile(
                self.pipe.unet, mode="reduce-overhead", fullgraph=True
            )
            self.pipe.vae = torch.compile(
                self.pipe.vae, mode="reduce-overhead", fullgraph=True
            )
            self.pipe(
                prompt="warmup",
                image=[Image.new("RGB", (768, 768))],
                control_image=[Image.new("RGB", (768, 768))],
            )

    def predict(self, params: "Pipeline.InputParams") -> Image.Image:
        generator = torch.manual_seed(params.seed)

        prompt = params.prompt
        negative_prompt = params.negative_prompt
        prompt_embeds = None
        pooled_prompt_embeds = None
        negative_prompt_embeds = None
        negative_pooled_prompt_embeds = None
        if hasattr(self.pipe, "compel_proc"):
            _prompt_embeds, pooled_prompt_embeds = self.pipe.compel_proc(
                [params.prompt, params.negative_prompt]
            )
            prompt = None
            negative_prompt = None
            prompt_embeds = _prompt_embeds[0:1]
            pooled_prompt_embeds = pooled_prompt_embeds[0:1]
            negative_prompt_embeds = _prompt_embeds[1:2]
            negative_pooled_prompt_embeds = pooled_prompt_embeds[1:2]

        control_image = self.canny_torch(
            params.image, params.canny_low_threshold, params.canny_high_threshold
        )
        steps = params.steps
        strength = params.strength
        if int(steps * strength) < 1:
            steps = math.ceil(1 / max(0.10, strength))

        results = self.pipe(
            image=params.image,
            control_image=control_image,
            prompt=prompt,
            negative_prompt=negative_prompt,
            prompt_embeds=prompt_embeds,
            pooled_prompt_embeds=pooled_prompt_embeds,
            negative_prompt_embeds=negative_prompt_embeds,
            negative_pooled_prompt_embeds=negative_pooled_prompt_embeds,
            generator=generator,
            strength=strength,
            num_inference_steps=steps,
            guidance_scale=params.guidance_scale,
            width=params.width,
            height=params.height,
            output_type="pil",
            controlnet_conditioning_scale=params.controlnet_scale,
            control_guidance_start=params.controlnet_start,
            control_guidance_end=params.controlnet_end,
        )

        nsfw_content_detected = (
            results.nsfw_content_detected[0]
            if "nsfw_content_detected" in results
            else False
        )
        if nsfw_content_detected:
            return None
        result_image = results.images[0]
        if params.debug_canny:
            # paste control_image on top of result_image
            w0, h0 = (200, 200)
            control_image = control_image.resize((w0, h0))
            w1, h1 = result_image.size
            result_image.paste(control_image, (w1 - w0, h1 - h0))

        return result_image