Merge branch 'beta' into scanner

2024-11-25 17:55:09 +01:00 · 2022-11-18 16:01:50 +05:30 · 2022-11-18 16:01:50 +05:30 · 1f815d7562
commit 1f815d7562
parent 7b4cfbeeaa f25d35fad8
9 changed files with 199 additions and 49 deletions
--- a/CHANGES.md
+++ b/CHANGES.md
@ -1,10 +1,30 @@
 # What's new?

-### 2.4.6
-* 16 Nov 2022 - Fix a regression in VRAM usage during startup, which caused 'Out of Memory' errors when starting on GPUs with 4gb (or less) VRAM
-* 16 Nov 2022 - Add Picklescan - a scanner for malicious model files. If it finds a malicious file, it will halt the web application and alert the user
+## v2.4
+### Major Changes
+- **Support for custom VAE models**. You can place your VAE files in the `models/vae` folder, and refresh the browser page to use them. More info: https://github.com/cmdr2/stable-diffusion-ui/wiki/VAE-Variational-Auto-Encoder
+- **Experimental support for multiple GPUs!** It should work automatically. Just open one browser tab per GPU, and spread your tasks across your GPUs. For e.g. open our UI in two browser tabs if you have two GPUs. You can customize which GPUs it should use in the "Settings" tab, otherwise let it automatically pick the best GPUs. Thanks @madrang . More info: https://github.com/cmdr2/stable-diffusion-ui/wiki/Run-on-Multiple-GPUs
+- **Cleaner UI design** - Show settings and help in new tabs, instead of dropdown popups (which were buggy). Thanks @mdiller
+- **Progress bar.** Thanks @mdiller
+- **Custom Image Modifiers** - You can now save your custom image modifiers! Your saved modifiers can include special characters like `{}, (), [], |`
+- Drag and Drop **text files generated from previously saved images**, and copy settings to clipboard. Thanks @madrang
+- Paste settings from clipboard. Thanks @JeLuf
+- Bug fixes to reduce the chances of tasks crashing during long multi-hour runs (chrome can put long-running background tabs to sleep). Thanks @JeLuf and @madrang
+- **Improved documentation.** Thanks @JeLuf and @jsuelwald
+- Improved the codebase for dealing with system settings and UI settings. Thanks @mdiller
+- Help instructions next to some setttings, and in the tab
+- Show system info in the settings tab
+- Keyboard shortcut: Ctrl+Enter to start a task
+- Configuration to prevent the browser from opening on startup
+- Lots of minor bug fixes
+- A `What's New?` tab in the UI

-### 2.4.5
-* 16 Nov 2022 - Add checkbox for "Open browser on startup".
-* 16 Nov 2022 - Add a directory for core plugins that ship with Stable Diffusion UI by default.
-* 16 Nov 2022 - Add a "What's New?" tab as a core plugin, which fetches the contents of CHANGES.md from the app's release branch.
+### Detailed changelog
+* 2.4.9 - 18 Nov 2022 - Add Picklescan - a scanner for malicious model files. If it finds a malicious file, it will halt the web application and alert the user. Thanks @JeLuf
+* 2.4.8 - 18 Nov 2022 - A `Use as Input` button to use the settings from a previously generated image task. Thanks @patriceac
+* 2.4.7 - 18 Nov 2022 - Don't crash if a VAE file fails to load
+* 2.4.7 - 17 Nov 2022 - Fix a bug where Face Correction (GFPGAN) would fail on cuda:N (i.e. GPUs other than cuda:0), as well as fail on CPU if the system had an incompatible GPU.
+* 2.4.6 - 16 Nov 2022 - Fix a regression in VRAM usage during startup, which caused 'Out of Memory' errors when starting on GPUs with 4gb (or less) VRAM
+* 2.4.5 - 16 Nov 2022 - Add checkbox for "Open browser on startup".
+* 2.4.5 - 16 Nov 2022 - Add a directory for core plugins that ship with Stable Diffusion UI by default.
+* 2.4.5 - 16 Nov 2022 - Add a "What's New?" tab as a core plugin, which fetches the contents of CHANGES.md from the app's release branch.
--- a/ui/index.html
+++ b/ui/index.html
@ -7,7 +7,7 @@
    <link rel="icon" type="image/png" href="/media/images/favicon-32x32.png" sizes="32x32">
    <link rel="stylesheet" href="/media/css/fonts.css?v=1">
    <link rel="stylesheet" href="/media/css/themes.css?v=3">
-    <link rel="stylesheet" href="/media/css/main.css?v=17">
+    <link rel="stylesheet" href="/media/css/main.css?v=22">
    <link rel="stylesheet" href="/media/css/auto-save.css?v=5">
    <link rel="stylesheet" href="/media/css/modifier-thumbnails.css?v=4">
    <link rel="stylesheet" href="/media/css/fontawesome-all.min.css?v=1">
@ -20,7 +20,7 @@
 <div id="container">
    <div id="top-nav">
        <div id="logo">
-            <h1>Stable Diffusion UI <small>v2.4.6 <span id="updateBranchLabel"></span></small></h1>
+            <h1>Stable Diffusion UI <small>v2.4.8 <span id="updateBranchLabel"></span></small></h1>
        </div>
        <div id="server-status">
            <div id="server-status-color">●</div>
@ -264,6 +264,7 @@
                                <li> <a href="https://github.com/cmdr2/stable-diffusion-ui/wiki/UI-Overview" target="_blank"><i class="fa-solid fa-list fa-fw"></i> UI Overview</a>
                                <li> <a href="https://github.com/cmdr2/stable-diffusion-ui/wiki/Writing-Prompts" target="_blank"><i class="fa-solid fa-pen-to-square fa-fw"></i> Writing prompts</a>
                                <li> <a href="https://github.com/cmdr2/stable-diffusion-ui/wiki/Inpainting" target="_blank"><i class="fa-solid fa-paintbrush fa-fw"></i> Inpainting</a>
+                                <li> <a href="https://github.com/cmdr2/stable-diffusion-ui/wiki/Run-on-Multiple-GPUs" target="_blank"><i class="fa-solid fa-paintbrush fa-fw"></i> Run on Multiple GPUs</a>
                            </ul>
  
                            <li><span class="help-section">Installation</span>
@ -328,15 +329,15 @@
 </div>
 </body>

-<script src="media/js/parameters.js?v=9"></script>
-<script src="media/js/plugins.js?v=1"></script>
 <script src="media/js/utils.js?v=6"></script>
+<script src="media/js/parameters.js?v=10"></script>
+<script src="media/js/plugins.js?v=1"></script>
 <script src="media/js/inpainting-editor.js?v=1"></script>
-<script src="media/js/image-modifiers.js?v=6"></script>
+<script src="media/js/image-modifiers.js?v=7"></script>
 <script src="media/js/auto-save.js?v=8"></script>
-<script src="media/js/main.js?v=23"></script>
+<script src="media/js/main.js?v=26"></script>
 <script src="media/js/themes.js?v=4"></script>
-<script src="media/js/dnd.js?v=9"></script>
+<script src="media/js/dnd.js?v=11"></script>
 <script>
 async function init() {
    await initSettings()
--- a/ui/media/css/main.css
+++ b/ui/media/css/main.css
@ -22,6 +22,11 @@ a:visited {
 label {
    font-size: 10pt;
 }
+code {
+    background: var(--background-color4);
+    padding: 2px 4px;
+    border-radius: 4px;
+}
 #prompt {
    width: 100%;
    height: 65pt;
@ -437,6 +442,17 @@ img {
 .secondaryButton:hover {
    background: rgb(177, 27, 0);
 }
+.useSettings {
+    background: var(--accent-color);
+    border: 1px solid var(--accent-color);
+    color: rgb(255, 221, 255);
+    padding: 3pt 6pt;
+    margin-right: 6pt;
+    float: right;
+}
+.useSettings:hover {
+    background: hsl(var(--accent-hue), 100%, calc(var(--accent-lightness) + 6%));
+}
 .stopTask {
    float: right;
 }
@ -657,7 +673,7 @@ input::file-selector-button {
@media (min-width: 700px) {
    /* #editor {
        max-width: 480px;
-    } */
+    }*/
    .float-container {
        padding: 20px;
    }
@ -898,6 +914,9 @@ input::file-selector-button {
 i.active {
    background: var(--accent-color);
 }
+.primaryButton.active {
+    background: hsl(var(--accent-hue), 100%, 50%);
+}
 #system-info {
    max-width: 800px;
    font-size: 10pt;
--- a/ui/media/js/dnd.js
+++ b/ui/media/js/dnd.js
@ -243,7 +243,9 @@ const TASK_MAPPING = {
        parse: (val) => val
    }
 }
-function restoreTaskToUI(task) {
+function restoreTaskToUI(task, fieldsToSkip) {
+    fieldsToSkip = fieldsToSkip || []
+
    if ('numOutputsTotal' in task) {
        numOutputsTotalField.value = task.numOutputsTotal
    }
@ -255,10 +257,47 @@ function restoreTaskToUI(task) {
        return
    }
    for (const key in TASK_MAPPING) {
-        if (key in task.reqBody) {
+        if (key in task.reqBody && !fieldsToSkip.includes(key)) {
            TASK_MAPPING[key].setUI(task.reqBody[key])
        }
    }
+
+    // restore the original tag
+    promptField.value = task.reqBody.original_prompt || task.reqBody.prompt
+
+    // Restore modifiers
+    if (task.reqBody.active_tags) {
+        refreshModifiersState(task.reqBody.active_tags)
+    }
+
+    // properly reset checkboxes
+    if (!('use_face_correction' in task.reqBody)) {
+        useFaceCorrectionField.checked = false
+    }
+    if (!('use_upscale' in task.reqBody)) {
+        useUpscalingField.checked = false
+    }
+    if (!('mask' in task.reqBody)) {
+        maskSetting.checked = false
+    }
+    upscaleModelField.disabled = !useUpscalingField.checked
+
+    // Show the source picture if present
+    initImagePreview.src = (task.reqBody.init_image == undefined ? '' : task.reqBody.init_image)
+    if (IMAGE_REGEX.test(initImagePreview.src)) {
+        Boolean(task.reqBody.mask) ? inpaintingEditor.setImg(task.reqBody.mask) : inpaintingEditor.resetBackground()
+        initImagePreviewContainer.style.display = 'block'
+        inpaintingEditorContainer.style.display = 'none'
+        promptStrengthContainer.style.display = 'table-row'
+        //samplerSelectionContainer.style.display = 'none'
+        // maskSetting.checked = false
+        inpaintingEditorContainer.style.display = maskSetting.checked ? 'block' : 'none'
+    } else {
+        initImagePreviewContainer.style.display = 'none'
+        // inpaintingEditorContainer.style.display = 'none'
+        promptStrengthContainer.style.display = 'none'
+        // maskSetting.style.display = 'none'
+    }
 }
 function readUI() {
    const reqBody = {}
@ -430,10 +469,10 @@ function checkWriteToClipboardPermission (result) {
        copyIcon.innerHTML = `<span class="simple-tooltip right">Copy Image Settings</span>`
        copyIcon.addEventListener('click', (event) => {
            event.stopPropagation()
-	    // Add css class 'active'
-	    copyIcon.classList.add('active')
-	    // In 1000 ms remove the 'active' class
-	    asyncDelay(1000).then(() => copyIcon.classList.remove('active'))
+            // Add css class 'active'
+            copyIcon.classList.add('active')
+            // In 350 ms remove the 'active' class
+            asyncDelay(350).then(() => copyIcon.classList.remove('active'))
            const uiState = readUI()
            TASK_REQ_NO_EXPORT.forEach((key) => delete uiState.reqBody[key])
            if (uiState.reqBody.init_image && !IMAGE_REGEX.test(uiState.reqBody.init_image)) {
@ -450,10 +489,10 @@ function checkWriteToClipboardPermission (result) {
        pasteIcon.innerHTML = `<span class="simple-tooltip right">Paste Image Settings</span>`
        pasteIcon.addEventListener('click', (event) => {
            event.stopPropagation()
-	    // Add css class 'active'
-	    pasteIcon.classList.add('active')
-	    // In 1000 ms remove the 'active' class
-	    asyncDelay(1000).then(() => pasteIcon.classList.remove('active'))
+            // Add css class 'active'
+            pasteIcon.classList.add('active')
+            // In 350 ms remove the 'active' class
+            asyncDelay(350).then(() => pasteIcon.classList.remove('active'))
            pasteFromClipboard()
        })
        resetSettings.parentNode.insertBefore(pasteIcon, resetSettings)
--- a/ui/media/js/image-modifiers.js
+++ b/ui/media/js/image-modifiers.js
@ -148,6 +148,58 @@ async function loadModifiers() {
    loadCustomModifiers()
 }

+function refreshModifiersState(newTags) {
+    // clear existing modifiers
+    document.querySelector('#editor-modifiers').querySelectorAll('.modifier-card').forEach(modifierCard => {
+        const modifierName = modifierCard.querySelector('.modifier-card-label').innerText
+        if (activeTags.map(x => x.name).includes(modifierName)) {
+            modifierCard.classList.remove(activeCardClass)
+            modifierCard.querySelector('.modifier-card-image-overlay').innerText = '+'
+        }
+    })
+    activeTags = []
+
+    // set new modifiers
+    newTags.forEach(tag => {
+        let found = false
+        document.querySelector('#editor-modifiers').querySelectorAll('.modifier-card').forEach(modifierCard => {
+            const modifierName = modifierCard.querySelector('.modifier-card-label').innerText
+            if (tag == modifierName) {
+                // add modifier to active array
+                activeTags.push({
+                    'name': modifierName,
+                    'element': modifierCard.cloneNode(true),
+                    'originElement': modifierCard
+                })
+                modifierCard.classList.add(activeCardClass)
+                modifierCard.querySelector('.modifier-card-image-overlay').innerText = '-'
+                found = true
+            }
+        })
+        if (found == false) { // custom tag went missing, create one here
+            let modifierCard = createModifierCard(tag, undefined) // create a modifier card for the missing tag, no image
+            
+            modifierCard.addEventListener('click', () => {
+                if (activeTags.map(x => x.name).includes(tag)) {
+                    // remove modifier from active array
+                    activeTags = activeTags.filter(x => x.name != tag)
+                    modifierCard.classList.remove(activeCardClass)
+
+                    modifierCard.querySelector('.modifier-card-image-overlay').innerText = '+'
+                }
+                refreshTagsList()
+            })
+
+            activeTags.push({
+                'name': tag,
+                'element': modifierCard,
+                'originElement': undefined  // no origin element for missing tags
+            })
+        }
+    })
+    refreshTagsList()
+}
+
 function refreshTagsList() {
    editorModifierTagsList.innerHTML = ''

@ -167,7 +219,7 @@ function refreshTagsList() {
        tag.element.addEventListener('click', () => {
            let idx = activeTags.indexOf(tag)

-            if (idx !== -1) {
+            if (idx !== -1 && activeTags[idx].originElement !== undefined) {
                activeTags[idx].originElement.classList.remove(activeCardClass)
                activeTags[idx].originElement.querySelector('.modifier-card-image-overlay').innerText = '+'

--- a/ui/media/js/main.js
+++ b/ui/media/js/main.js
@ -789,7 +789,9 @@ function getCurrentUserRequest() {
            stream_progress_updates: true,
            stream_image_progress: (numOutputsTotal > 50 ? false : streamImageProgressField.checked),
            show_only_filtered_image: showOnlyFilteredImageField.checked,
-            output_format: outputFormatField.value
+            output_format: outputFormatField.value,
+            original_prompt: promptField.value,
+            active_tags: (activeTags.map(x => x.name))
        }
    }
    if (IMAGE_REGEX.test(initImagePreview.src)) {
@ -856,6 +858,7 @@ function createTask(task) {
    taskEntry.innerHTML = ` <div class="header-content panel collapsible active">
                                <div class="taskStatusLabel">Enqueued</div>
                                <button class="secondaryButton stopTask"><i class="fa-solid fa-trash-can"></i> Remove</button>
+                                <button class="secondaryButton useSettings"><i class="fa-solid fa-redo"></i> Use as Input</button>
                                <div class="preview-prompt collapsible active"></div>
                                <div class="taskConfig">${taskConfig}</div>
                                <div class="outputMsg"></div>
@ -894,6 +897,12 @@ function createTask(task) {
        }
    })

+    task['useSettings'] = taskEntry.querySelector('.useSettings')
+    task['useSettings'].addEventListener('click', function(e) {
+        e.stopPropagation()
+        restoreTaskToUI(task, TASK_REQ_NO_EXPORT)
+    })
+
    imagePreview.insertBefore(taskEntry, previewTools.nextSibling)

    task.previewPrompt.innerText = task.reqBody.prompt
--- a/ui/media/js/parameters.js
+++ b/ui/media/js/parameters.js
@ -327,4 +327,7 @@ saveSettingsBtn.addEventListener('click', function() {
 		'update_branch': updateBranch,
 		'ui_open_browser_on_start': uiOpenBrowserOnStartField.checked
    })
+
+	saveSettingsBtn.classList.add('active')
+	asyncDelay(300).then(() => saveSettingsBtn.classList.remove('active'))
 })
--- a/ui/sd_internal/runtime.py
+++ b/ui/sd_internal/runtime.py
@ -140,15 +140,24 @@ def load_model_ckpt():
    _, _ = modelFS.load_state_dict(sd, strict=False)

    if thread_data.vae_file is not None:
-        for model_extension in ['.ckpt', '.vae.pt']:
-            if os.path.exists(thread_data.vae_file + model_extension):
-                print(f"Loading VAE weights from: {thread_data.vae_file}{model_extension}")
-                vae_ckpt = torch.load(thread_data.vae_file + model_extension, map_location="cpu")
-                vae_dict = {k: v for k, v in vae_ckpt["state_dict"].items() if k[0:4] != "loss"}
-                modelFS.first_stage_model.load_state_dict(vae_dict, strict=False)
-                break
-            else:
-                print(f'Cannot find VAE file: {thread_data.vae_file}{model_extension}')
+        try:
+            loaded = False
+            for model_extension in ['.ckpt', '.vae.pt']:
+                if os.path.exists(thread_data.vae_file + model_extension):
+                    print(f"Loading VAE weights from: {thread_data.vae_file}{model_extension}")
+                    vae_ckpt = torch.load(thread_data.vae_file + model_extension, map_location="cpu")
+                    vae_dict = {k: v for k, v in vae_ckpt["state_dict"].items() if k[0:4] != "loss"}
+                    modelFS.first_stage_model.load_state_dict(vae_dict, strict=False)
+                    loaded = True
+                    break
+
+            if not loaded:
+                print(f'Cannot find VAE: {thread_data.vae_file}')
+                thread_data.vae_file = None
+        except:
+            print(traceback.format_exc())
+            print(f'Could not load VAE: {thread_data.vae_file}')
+            thread_data.vae_file = None

    modelFS.eval()
    # if thread_data.device != 'cpu':
@ -236,9 +245,14 @@ def wait_model_move_to(model, target_device): # Send to target_device and wait u

 def load_model_gfpgan():
    if thread_data.gfpgan_file is None: raise ValueError(f'Thread gfpgan_file is undefined.')
+
+    # hack for a bug in facexlib: https://github.com/xinntao/facexlib/pull/19/files
+    from facexlib.detection import retinaface
+    retinaface.device = torch.device(thread_data.device)
+    print('forced retinaface.device to', thread_data.device)
+
    model_path = thread_data.gfpgan_file + ".pth"
-    device = 'cuda:0' if force_gfpgan_to_cuda0 else thread_data.device
-    thread_data.model_gfpgan = GFPGANer(device=torch.device(device), model_path=model_path, upscale=1, arch='clean', channel_multiplier=2, bg_upsampler=None)
+    thread_data.model_gfpgan = GFPGANer(device=torch.device(thread_data.device), model_path=model_path, upscale=1, arch='clean', channel_multiplier=2, bg_upsampler=None)
    print('loaded', thread_data.gfpgan_file, 'to', thread_data.model_gfpgan.device, 'precision', thread_data.precision)

 def load_model_real_esrgan():
@ -288,10 +302,10 @@ def apply_filters(filter_name, image_data, model_path=None):
    print(f'Applying filter {filter_name}...')
    gc() # Free space before loading new data.

-    if filter_name == 'gfpgan':
-        if isinstance(image_data, torch.Tensor):
-            image_data.to('cuda:0' if force_gfpgan_to_cuda0 else thread_data.device)
+    if isinstance(image_data, torch.Tensor):
+        image_data.to(thread_data.device)

+    if filter_name == 'gfpgan':
        if model_path is not None and model_path != thread_data.gfpgan_file:
            thread_data.gfpgan_file = model_path
            load_model_gfpgan()
@ -303,9 +317,6 @@ def apply_filters(filter_name, image_data, model_path=None):
        image_data = output[:,:,::-1]

    if filter_name == 'real_esrgan':
-        if isinstance(image_data, torch.Tensor):
-            image_data.to(thread_data.device)
-
        if model_path is not None and model_path != thread_data.real_esrgan_file:
            thread_data.real_esrgan_file = model_path
            load_model_real_esrgan()
--- a/ui/sd_internal/task_manager.py
+++ b/ui/sd_internal/task_manager.py
@ -217,10 +217,6 @@ def thread_get_next_task():
    task = None
    try:  # Select a render task.
        for queued_task in tasks_queue:
-            if queued_task.request.use_face_correction and runtime.thread_data.device == 'cpu' and is_alive() == 1:
-                queued_task.error = Exception('The CPU cannot be used to run this task currently. Please remove "Fix incorrect faces" from Image Settings and try again.')
-                task = queued_task
-                break
            if queued_task.render_device and runtime.thread_data.device != queued_task.render_device:
                # Is asking for a specific render device.
                if is_alive(queued_task.render_device) > 0:
@ -440,7 +436,7 @@ def stop_render_thread(device):
    try:
        device_manager.validate_device_id(device, log_prefix='stop_render_thread')
    except:
-        print(traceback.format_exec())
+        print(traceback.format_exc())
        return False

    if not manager_lock.acquire(blocking=True, timeout=LOCK_TIMEOUT): raise Exception('stop_render_thread' + ERR_LOCK_FAILED)