Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
| """The model used in this Space alters the underlying Stable Diffusion model available at | |
| https://huggingface.co/CompVis/stable-diffusion-v1-4 through the addition of new embedding vectors | |
| in order to capture the likeness of the Determined AI logo. These alternations are fully captured | |
| in the learned_embeddings_dict.pt pickle file in the root of the repository.""" | |
| import pathlib | |
| import os | |
| from PIL import Image | |
| import gradio as gr | |
| import torch | |
| from diffusers import StableDiffusionPipeline | |
| import utils | |
| use_auth_token = os.environ["HF_AUTH_TOKEN"] | |
| NSFW_IMAGE = Image.open("nsfw.png") | |
| BATCH_SIZE = 2 | |
| # Instantiate the pipeline. | |
| device, revision, torch_dtype = ( | |
| ("cuda", "fp16", torch.float16) | |
| if torch.cuda.is_available() | |
| else ("cpu", "main", torch.float32) | |
| ) | |
| pipeline = StableDiffusionPipeline.from_pretrained( | |
| pretrained_model_name_or_path="CompVis/stable-diffusion-v1-4", | |
| use_auth_token=use_auth_token, | |
| revision=revision, | |
| torch_dtype=torch_dtype, | |
| ).to(device) | |
| # Load in the new concepts. | |
| CONCEPT_PATH = pathlib.Path("learned_embeddings_dict.pt") | |
| learned_embeddings_dict = torch.load(CONCEPT_PATH) | |
| concept_to_dummy_strs_map = {} | |
| for concept_token, embedding_dict in learned_embeddings_dict.items(): | |
| initializer_strs = embedding_dict["initializer_strs"] | |
| learned_embeddings = embedding_dict["learned_embeddings"] | |
| ( | |
| initializer_ids, | |
| dummy_placeholder_ids, | |
| dummy_placeholder_strs, | |
| ) = utils.add_new_tokens_to_tokenizer( | |
| concept_str=concept_token, | |
| initializer_strs=initializer_strs, | |
| tokenizer=pipeline.tokenizer, | |
| ) | |
| pipeline.text_encoder.resize_token_embeddings(len(pipeline.tokenizer)) | |
| token_embeddings = pipeline.text_encoder.get_input_embeddings().weight.data | |
| for d_id, tensor in zip(dummy_placeholder_ids, learned_embeddings): | |
| token_embeddings[d_id] = tensor | |
| concept_to_dummy_strs_map[concept_token] = dummy_placeholder_strs | |
| def replace_concept_strs(text: str): | |
| for concept_token, dummy_strs in concept_to_dummy_strs_map.items(): | |
| text = text.replace(concept_token, dummy_strs) | |
| return text | |
| def inference(prompt: str, guidance_scale: int, num_inference_steps: int, seed: int): | |
| if not prompt: | |
| raise ValueError("Please enter a prompt.") | |
| if 'det-logo' not in prompt: | |
| raise ValueError('"det-logo" must be included in the prompt.') | |
| prompt = replace_concept_strs(prompt) | |
| generator = torch.Generator(device=device).manual_seed(seed) | |
| output = pipeline( | |
| prompt=[prompt] * BATCH_SIZE, | |
| num_inference_steps=num_inference_steps, | |
| guidance_scale=guidance_scale, | |
| generator=generator, | |
| ) | |
| img_list, nsfw_list = output.images, output.nsfw_content_detected | |
| filtered_imgs = [ | |
| img if not nsfw else NSFW_IMAGE for img, nsfw in zip(img_list, nsfw_list) | |
| ] | |
| return filtered_imgs | |
| css = """ | |
| .gradio-container { | |
| font-family: 'Roboto', sans-serif; background-color: white !important; font-color: black !important; | |
| } | |
| .flex-grow { | |
| font-family: 'Roboto', sans-serif; background-color: white !important; font-color: black !important; | |
| } | |
| .font-mono { | |
| font-family: 'Roboto', sans-serif; background-color: white !important; font-color: black !important; | |
| } | |
| .gr-padded { | |
| font-family: 'Roboto', sans-serif; background-color: white !important; font-color: black !important; color: black !important; | |
| } | |
| .bg-gray-700 { | |
| font-family: 'Roboto', sans-serif; background-color: white !important; font-color: black !important; color: white !important; | |
| } | |
| .gr-box { | |
| font-family: 'Roboto', sans-serif; background-color: white !important; font-color: black !important; color: black !important; | |
| } | |
| .h-6 { | |
| font-family: 'Roboto', sans-serif; background-color: white !important; font-color: black !important; | |
| } | |
| .h-6 { | |
| font-family: 'Roboto', sans-serif; background-color: white !important; font-color: black !important; | |
| } | |
| .gr-samples-gallery { | |
| font-family: 'Roboto', sans-serif; background-color: white !important; font-color: black !important; color: black !important; | |
| } | |
| .gr-sample-textbox:hover { | |
| font-family: 'Roboto', sans-serif; background-color: #BAD7DF !important; font-color: black !important; color: black !important; | |
| } | |
| h1 { | |
| font-family: 'Roboto', sans-serif; color: black !important; | |
| } | |
| .text-gray-500 { | |
| font-family: 'Roboto', sans-serif; color: black !important; | |
| } | |
| .gr-button { | |
| color: white !important; | |
| border-color: black; | |
| background: white !important; | |
| } | |
| .flex-wrap { | |
| color: white !important; | |
| border-color: white !important; | |
| background: white !important; | |
| } | |
| .grow-0 { | |
| color: black !important; | |
| border-color: black; | |
| background: white !important; | |
| } | |
| .grow-0:hover { | |
| color: black !important; | |
| border-color: black; | |
| background: #BAD7DF !important; | |
| } | |
| input[type='range'] { | |
| accent-color: white; | |
| } | |
| .dark input[type='range'] { | |
| accent-color: #dfdfdf; | |
| } | |
| .container { | |
| max-width: 730px; | |
| margin: auto; | |
| padding-top: 1.5rem; | |
| background: white; | |
| } | |
| #gallery { | |
| margin-bottom: 1rem; | |
| margin-left: auto; | |
| margin-right: auto; | |
| border-bottom-right-radius: .5rem !important; | |
| border-bottom-left-radius: .5rem !important; | |
| } | |
| #gallery>div>.h-full { | |
| min-height: 20rem; | |
| } | |
| .details:hover { | |
| text-decoration: underline; | |
| } | |
| .gr-button { | |
| white-space: nowrap; | |
| } | |
| .gr-button:focus { | |
| border-color: rgb(147 197 253 / var(--tw-border-opacity)); | |
| outline: none; | |
| box-shadow: var(--tw-ring-offset-shadow), var(--tw-ring-shadow), var(--tw-shadow, 0 0 #0000); | |
| --tw-border-opacity: 1; | |
| --tw-ring-offset-shadow: var(--tw-ring-inset) 0 0 0 var(--tw-ring-offset-width) var(--tw-ring-offset-color); | |
| --tw-ring-shadow: var(--tw-ring-inset) 0 0 0 calc(3px var(--tw-ring-offset-width)) var(--tw-ring-color); | |
| --tw-ring-color: rgb(191 219 254 / var(--tw-ring-opacity)); | |
| --tw-ring-opacity: .5; | |
| } | |
| #advanced-btn, #license-btn { | |
| font-size: .7rem !important; | |
| line-height: 19px; | |
| margin-top: 12px; | |
| margin-bottom: 12px; | |
| padding: 2px 8px; | |
| border-radius: 14px !important; | |
| background: white !important; | |
| color: black !important; | |
| } | |
| #license-btn:hover { | |
| color: black !important; | |
| border-color: black; | |
| background: #BAD7DF !important; | |
| } | |
| #advanced-btn:hover { | |
| color: black !important; | |
| border-color: black; | |
| background: #BAD7DF !important; | |
| } | |
| #advanced-option`s { | |
| display: none; | |
| margin-bottom: 20px; | |
| } | |
| #license-display { | |
| display: none; | |
| margin-bottom: 20px; | |
| } | |
| #component-1 { | |
| max-height: 3rem; | |
| margin-bottom: 1rem; | |
| margin-left: auto; | |
| margin-right: auto; | |
| border-bottom-right-radius: .5rem !important; | |
| border-bottom-left-radius: .5rem !important; | |
| } | |
| #component-21 { | |
| max-height: 2rem; | |
| } | |
| .footer { | |
| margin-bottom: 0px; | |
| margin-top: 0px; | |
| text-align: center; | |
| border-bottom: 1px solid #e5e5e5; | |
| background: white !important; | |
| color: black !important; | |
| } | |
| .footer>p { | |
| font-size: .8rem; | |
| display: inline-block; | |
| padding: 0 10px; | |
| transform: translateY(10px); | |
| background: white !important; | |
| } | |
| .dark .footer { | |
| border-color: #303030; | |
| } | |
| .dark .footer>p { | |
| background: #0b0f19; | |
| } | |
| .acknowledgments h4{ | |
| margin: 1.25em 0 .25em 0; | |
| font-weight: bold; | |
| font-size: 115%; | |
| } | |
| #container-advanced-btns{ | |
| display: flex; | |
| flex-wrap: wrap; | |
| justify-content: space-between; | |
| align-items: center; | |
| } | |
| #container-license-btns{ | |
| margin: 1.25em 0 .25em 0; | |
| display: flex; | |
| flex-wrap: wrap; | |
| justify-content: space-between; | |
| align-items: center; | |
| } | |
| .animate-spin { | |
| animation: spin 1s linear infinite; | |
| } | |
| @keyframes spin { | |
| from { | |
| transform: rotate(0deg); | |
| } | |
| to { | |
| transform: rotate(360deg); | |
| } | |
| } | |
| .gr-form{ | |
| flex: 1 1 50%; border-top-right-radius: 0; border-bottom-right-radius: 0; | |
| } | |
| #prompt-container{ | |
| gap: 0; | |
| } | |
| """ | |
| block = gr.Blocks(css=css) | |
| examples = [ | |
| [ | |
| "A surrealist oil painting by Salvador Dali of a det-logo using soft, blended colors", | |
| # 4, | |
| # 45, | |
| # 7.5, | |
| # 1024, | |
| ], | |
| [ | |
| "Beautiful tarot illustration of a det-logo, in the style of james jean and victo ngai, mystical colors, trending on artstation", | |
| # 4, | |
| # 45,` | |
| # 7, | |
| # 1024, | |
| ], | |
| [ | |
| "Black and white ink doodle illustration of an overgrown det-logo, style by peter deligdisch, peterdraws", | |
| # 4, | |
| # 45, | |
| # 7, | |
| # 1024, | |
| ], | |
| ] | |
| with block: | |
| gr.HTML( | |
| """ | |
| <div style="text-align: center; max-width: 650px; margin: 1.25em 0 .25em 0;"> | |
| <div | |
| style=" | |
| display: inline-flex; | |
| align-items: center; | |
| font-size: 1.5rem; | |
| " | |
| > | |
| <h1 style="font-weight: 600;"> | |
| Determined AI Textual Inversion Demo | |
| </h1> | |
| </div> | |
| </div> | |
| """ | |
| ) | |
| with gr.Group(): | |
| with gr.Box(): | |
| with gr.Row(elem_id="prompt-container").style(equal_height=True): | |
| prompt = gr.Textbox( | |
| label='Enter a prompt including "det-logo"', | |
| show_label=False, | |
| max_lines=1, | |
| placeholder='Enter a prompt including "det-logo"', | |
| elem_id="prompt-text-input", | |
| ).style( | |
| container=False, | |
| ) | |
| btn = gr.Button("Generate image").style( | |
| full_width=False, | |
| ) | |
| gallery = gr.Gallery( | |
| label="Generated images", show_label=False, elem_id="gallery" | |
| ).style(grid=[BATCH_SIZE], height="auto") | |
| with gr.Group(elem_id="container-advanced-btns"): | |
| advanced_button = gr.Button("Advanced options", elem_id="advanced-btn") | |
| with gr.Row(elem_id="advanced-options"): | |
| num_inference_steps = gr.Slider( | |
| label="Steps", minimum=1, maximum=80, value=50, step=1 | |
| ) | |
| guidance_scale = gr.Slider( | |
| label="Guidance Scale", minimum=1.0, maximum=25.0, value=7.5, step=0.1 | |
| ) | |
| seed = gr.Slider( | |
| label="Seed", | |
| minimum=0, | |
| maximum=2147483647, | |
| step=1, | |
| randomize=True, | |
| ) | |
| ex = gr.Examples( | |
| examples=examples, | |
| fn=inference, | |
| inputs=[prompt, guidance_scale, num_inference_steps, seed], | |
| outputs=[gallery], | |
| cache_examples=False, | |
| ) | |
| ex.dataset.headers = [""] | |
| prompt.submit( | |
| inference, | |
| inputs=[prompt, guidance_scale, num_inference_steps, seed], | |
| outputs=[gallery], | |
| ) | |
| btn.click( | |
| inference, | |
| inputs=[prompt, guidance_scale, num_inference_steps, seed], | |
| outputs=[gallery], | |
| ) | |
| advanced_button.click( | |
| None, | |
| [], | |
| prompt, | |
| _js=""" | |
| () => { | |
| var appDom = document.querySelector("body > gradio-app"); | |
| var options = appDom.querySelector("#advanced-options") | |
| if (options == null) {options = appDom.shadowRoot.querySelector("#advanced-options")} | |
| options.style.display = ["none", ""].includes(options.style.display) ? "flex" : "none"; | |
| }""", | |
| ) | |
| with gr.Group(elem_id="container-license-btns"): | |
| license_button = gr.Button("License, biases, and model changes", | |
| elem_id="license-btn") | |
| license_button.click( | |
| None, | |
| [], | |
| prompt, | |
| _js=""" | |
| () => { | |
| var appDom = document.querySelector("body > gradio-app"); | |
| var options = appDom.querySelector("#license-display") | |
| if (options == null) {options = appDom.shadowRoot.querySelector("#license-display")} | |
| options.style.display = ["none", ""].includes(options.style.display) ? "flex" : "none"; | |
| }""", | |
| ) | |
| with gr.Row(elem_id="license-display"): | |
| gr.HTML( | |
| """ | |
| <div class="acknowledgments"> | |
| <p><h4>LICENSE</h4> | |
| The model is licensed with a <a href="https://huggingface.co/spaces/CompVis/stable-diffusion-license" style="text-decoration: underline;" target="_blank">CreativeML Open RAIL-M</a> license. The authors claim no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in this license. The license forbids you from sharing any content that violates any laws, produce any harm to a person, disseminate any personal information that would be meant for harm, spread misinformation and target vulnerable groups. For the full list of restrictions please <a href="https://huggingface.co/spaces/CompVis/stable-diffusion-license" target="_blank" style="text-decoration: underline;" target="_blank">read the license</a></p> | |
| <p><h4>Biases and content acknowledgment</h4> | |
| Despite how impressive being able to turn text into image is, beware to the fact that this model may output content that reinforces or exacerbates societal biases, as well as realistic faces, pornography and violence. The model was trained on the <a href="https://laion.ai/blog/laion-5b/" style="text-decoration: underline;" target="_blank">LAION-5B dataset</a>, which scraped non-curated image-text-pairs from the internet (the exception being the removal of illegal content) and is meant for research purposes. You can read more in the <a href="https://huggingface.co/CompVis/stable-diffusion-v1-4" style="text-decoration: underline;" target="_blank">model card</a></p> | |
| <p><h4>Model Changes</h4> | |
| The model used in this Space alters the underlying <a href="https://huggingface.co/CompVis/stable-diffusion-v1-4" style="text-decoration: underline;" target="_blank">stable-diffusion-v1-4</a> model through the addition of new embedding vectors in order to capture the likeness of the <a href="https://www.determined.ai" style="text-decoration: underline;" target="_blank">Determined AI</a> logo.</p> | |
| </div> | |
| """ | |
| ) | |
| gr.HTML( | |
| """ | |
| <div class="footer"> | |
| <p>Underlying model by <a href="https://huggingface.co/CompVis" style="text-decoration: underline;" target="_blank">CompVis</a> and <a href="https://huggingface.co/stabilityai" style="text-decoration: underline;" target="_blank">Stability AI</a> - Gradio code based on the <a href="https://huggingface.co/spaces/stabilityai/stable-diffusion" style="text-decoration: underline;" target="_blank">Stability AI Demo</a> | |
| </p> | |
| </div> | |
| """ | |
| ) | |
| block.queue(max_size=10).launch(show_error=True) | |