moved deepdanbooru to pure pytorch implementation

README.md

@@ -70,7 +70,7 @@ Check the [custom scripts](https://github.com/AUTOMATIC1111/stable-diffusion-web
      - separate prompts using uppercase `AND`
      - also supports weights for prompts: `a cat :1.2 AND a dog AND a penguin :2.2`
 - No token limit for prompts (original stable diffusion lets you use up to 75 tokens)
-- DeepDanbooru integration, creates danbooru style tags for anime prompts (add --deepdanbooru to commandline args)
+- DeepDanbooru integration, creates danbooru style tags for anime prompts
 - [xformers](https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Xformers), major speed increase for select cards: (add --xformers to commandline args)
 - via extension: [History tab](https://github.com/yfszzx/stable-diffusion-webui-images-browser): view, direct and delete images conveniently within the UI
 - Generate forever option

launch.py

@@ -134,7 +134,6 @@ def prepare_enviroment():
 
     gfpgan_package = os.environ.get('GFPGAN_PACKAGE', "git+https://github.com/TencentARC/GFPGAN.git@8d2447a2d918f8eba5a4a01463fd48e45126a379")
     clip_package = os.environ.get('CLIP_PACKAGE', "git+https://github.com/openai/CLIP.git@d50d76daa670286dd6cacf3bcd80b5e4823fc8e1")
-    deepdanbooru_package = os.environ.get('DEEPDANBOORU_PACKAGE', "git+https://github.com/KichangKim/DeepDanbooru.git@d91a2963bf87c6a770d74894667e9ffa9f6de7ff")
 
     xformers_windows_package = os.environ.get('XFORMERS_WINDOWS_PACKAGE', 'https://github.com/C43H66N12O12S2/stable-diffusion-webui/releases/download/f/xformers-0.0.14.dev0-cp310-cp310-win_amd64.whl')
 
@@ -158,7 +157,6 @@ def prepare_enviroment():
     sys.argv, update_check = extract_arg(sys.argv, '--update-check')
     sys.argv, run_tests = extract_arg(sys.argv, '--tests')
     xformers = '--xformers' in sys.argv
-    deepdanbooru = '--deepdanbooru' in sys.argv
     ngrok = '--ngrok' in sys.argv
 
     try:
@@ -193,9 +191,6 @@ def prepare_enviroment():
         elif platform.system() == "Linux":
             run_pip("install xformers", "xformers")
 
-    if not is_installed("deepdanbooru") and deepdanbooru:
-        run_pip(f"install {deepdanbooru_package}#egg=deepdanbooru[tensorflow] tensorflow==2.10.0 tensorflow-io==0.27.0", "deepdanbooru")
-
     if not is_installed("pyngrok") and ngrok:
         run_pip("install pyngrok", "ngrok")
 

modules/api/api.py

@@ -9,7 +9,7 @@ from fastapi.security import HTTPBasic, HTTPBasicCredentials
 from secrets import compare_digest
 
 import modules.shared as shared
-from modules import sd_samplers
+from modules import sd_samplers, deepbooru
 from modules.api.models import *
 from modules.processing import StableDiffusionProcessingTxt2Img, StableDiffusionProcessingImg2Img, process_images
 from modules.extras import run_extras, run_pnginfo
@@ -18,9 +18,6 @@ from modules.sd_models import checkpoints_list
 from modules.realesrgan_model import get_realesrgan_models
 from typing import List
 
-if shared.cmd_opts.deepdanbooru:
-    from modules.deepbooru import get_deepbooru_tags
-
 def upscaler_to_index(name: str):
     try:
         return [x.name.lower() for x in shared.sd_upscalers].index(name.lower())
@@ -245,10 +242,7 @@ class Api:
             if interrogatereq.model == "clip":
                 processed = shared.interrogator.interrogate(img)
             elif interrogatereq.model == "deepdanbooru":
-                if shared.cmd_opts.deepdanbooru:
-                    processed = get_deepbooru_tags(img)
-                else:
-                    raise HTTPException(status_code=404, detail="Model not found. Add --deepdanbooru when launching for using the model.")
+                processed = deepbooru.model.tag(img)
             else:
                 raise HTTPException(status_code=404, detail="Model not found")
         

modules/deepbooru.py

-import os.path
-from concurrent.futures import ProcessPoolExecutor
-import multiprocessing
-import time
+import os
 import re
 
+import torch
+from PIL import Image
+import numpy as np
+
+from modules import modelloader, paths, deepbooru_model, devices, images, shared
+
 re_special = re.compile(r'([\\()])')
 
-def get_deepbooru_tags(pil_image):
-    """
-    This method is for running only one image at a time for simple use.  Used to the img2img interrogate.
-    """
-    from modules import shared  # prevents circular reference
-
-    try:
-        create_deepbooru_process(shared.opts.interrogate_deepbooru_score_threshold, create_deepbooru_opts())
-        return get_tags_from_process(pil_image)
-    finally:
-        release_process()
-
-
-OPT_INCLUDE_RANKS = "include_ranks"
-def create_deepbooru_opts():
-    from modules import shared
-
-    return {
-        "use_spaces": shared.opts.deepbooru_use_spaces,
-        "use_escape": shared.opts.deepbooru_escape,
-        "alpha_sort": shared.opts.deepbooru_sort_alpha,
-        OPT_INCLUDE_RANKS: shared.opts.interrogate_return_ranks,
-    }
-
-
-def deepbooru_process(queue, deepbooru_process_return, threshold, deepbooru_opts):
-    model, tags = get_deepbooru_tags_model()
-    while True: # while process is running, keep monitoring queue for new image
-        pil_image = queue.get()
-        if pil_image == "QUIT":
-            break
-        else:
-            deepbooru_process_return["value"] = get_deepbooru_tags_from_model(model, tags, pil_image, threshold, deepbooru_opts)
-
-
-def create_deepbooru_process(threshold, deepbooru_opts):
-    """
-    Creates deepbooru process.  A queue is created to send images into the process.  This enables multiple images
-    to be processed in a row without reloading the model or creating a new process.  To return the data, a shared
-    dictionary is created to hold the tags created.  To wait for tags to be returned, a value of -1 is assigned
-    to the dictionary and the method adding the image to the queue should wait for this value to be updated with
-    the tags.
-    """
-    from modules import shared  # prevents circular reference
-    context = multiprocessing.get_context("spawn")
-    shared.deepbooru_process_manager = context.Manager()
-    shared.deepbooru_process_queue = shared.deepbooru_process_manager.Queue()
-    shared.deepbooru_process_return = shared.deepbooru_process_manager.dict()
-    shared.deepbooru_process_return["value"] = -1
-    shared.deepbooru_process = context.Process(target=deepbooru_process, args=(shared.deepbooru_process_queue, shared.deepbooru_process_return, threshold, deepbooru_opts))
-    shared.deepbooru_process.start()
-
-
-def get_tags_from_process(image):
-    from modules import shared
-
-    shared.deepbooru_process_return["value"] = -1
-    shared.deepbooru_process_queue.put(image)
-    while shared.deepbooru_process_return["value"] == -1:
-        time.sleep(0.2)
-    caption = shared.deepbooru_process_return["value"]
-    shared.deepbooru_process_return["value"] = -1
-
-    return caption
-
-
-def release_process():
-    """
-    Stops the deepbooru process to return used memory
-    """
-    from modules import shared  # prevents circular reference
-    shared.deepbooru_process_queue.put("QUIT")
-    shared.deepbooru_process.join()
-    shared.deepbooru_process_queue = None
-    shared.deepbooru_process = None
-    shared.deepbooru_process_return = None
-    shared.deepbooru_process_manager = None
-
-def get_deepbooru_tags_model():
-    import deepdanbooru as dd
-    import tensorflow as tf
-    import numpy as np
-    this_folder = os.path.dirname(__file__)
-    model_path = os.path.abspath(os.path.join(this_folder, '..', 'models', 'deepbooru'))
-    if not os.path.exists(os.path.join(model_path, 'project.json')):
-        # there is no point importing these every time
-        import zipfile
-        from basicsr.utils.download_util import load_file_from_url
-        load_file_from_url(
-            r"https://github.com/KichangKim/DeepDanbooru/releases/download/v3-20211112-sgd-e28/deepdanbooru-v3-20211112-sgd-e28.zip",
-            model_path)
-        with zipfile.ZipFile(os.path.join(model_path, "deepdanbooru-v3-20211112-sgd-e28.zip"), "r") as zip_ref:
-            zip_ref.extractall(model_path)
-        os.remove(os.path.join(model_path, "deepdanbooru-v3-20211112-sgd-e28.zip"))
-
-    tags = dd.project.load_tags_from_project(model_path)
-    model = dd.project.load_model_from_project(
-        model_path, compile_model=False
-    )
-    return model, tags
-
-
-def get_deepbooru_tags_from_model(model, tags, pil_image, threshold, deepbooru_opts):
-    import deepdanbooru as dd
-    import tensorflow as tf
-    import numpy as np
-
-    alpha_sort = deepbooru_opts['alpha_sort']
-    use_spaces = deepbooru_opts['use_spaces']
-    use_escape = deepbooru_opts['use_escape']
-    include_ranks = deepbooru_opts['include_ranks']
-
-    width = model.input_shape[2]
-    height = model.input_shape[1]
-    image = np.array(pil_image)
-    image = tf.image.resize(
-        image,
-        size=(height, width),
-        method=tf.image.ResizeMethod.AREA,
-        preserve_aspect_ratio=True,
-    )
-    image = image.numpy()  # EagerTensor to np.array
-    image = dd.image.transform_and_pad_image(image, width, height)
-    image = image / 255.0
-    image_shape = image.shape
-    image = image.reshape((1, image_shape[0], image_shape[1], image_shape[2]))
-
-    y = model.predict(image)[0]
-
-    result_dict = {}
-
-    for i, tag in enumerate(tags):
-        result_dict[tag] = y[i]
-
-    unsorted_tags_in_theshold = []
-    result_tags_print = []
-    for tag in tags:
-        if result_dict[tag] >= threshold:
+
+class DeepDanbooru:
+    def __init__(self):
+        self.model = None
+
+    def load(self):
+        if self.model is not None:
+            return
+
+        files = modelloader.load_models(
+            model_path=os.path.join(paths.models_path, "torch_deepdanbooru"),
+            model_url='https://github.com/AUTOMATIC1111/TorchDeepDanbooru/releases/download/v1/model-resnet_custom_v3.pt',
+            ext_filter=".pt",
+            download_name='model-resnet_custom_v3.pt',
+        )
+
+        self.model = deepbooru_model.DeepDanbooruModel()
+        self.model.load_state_dict(torch.load(files[0], map_location="cpu"))
+
+        self.model.eval()
+        self.model.to(devices.cpu, devices.dtype)
+
+    def start(self):
+        self.load()
+        self.model.to(devices.device)
+
+    def stop(self):
+        if not shared.opts.interrogate_keep_models_in_memory:
+            self.model.to(devices.cpu)
+            devices.torch_gc()
+
+    def tag(self, pil_image):
+        self.start()
+        res = self.tag_multi(pil_image)
+        self.stop()
+
+        return res
+
+    def tag_multi(self, pil_image, force_disable_ranks=False):
+        threshold = shared.opts.interrogate_deepbooru_score_threshold
+        use_spaces = shared.opts.deepbooru_use_spaces
+        use_escape = shared.opts.deepbooru_escape
+        alpha_sort = shared.opts.deepbooru_sort_alpha
+        include_ranks = shared.opts.interrogate_return_ranks and not force_disable_ranks
+
+        pic = images.resize_image(2, pil_image.convert("RGB"), 512, 512)
+        a = np.expand_dims(np.array(pic, dtype=np.float32), 0) / 255
+
+        with torch.no_grad(), devices.autocast():
+            x = torch.from_numpy(a).cuda()
+            y = self.model(x)[0].detach().cpu().numpy()
+
+        probability_dict = {}
+
+        for tag, probability in zip(self.model.tags, y):
+            if probability < threshold:
+                continue
+
             if tag.startswith("rating:"):
                 continue
-            unsorted_tags_in_theshold.append((result_dict[tag], tag))
-            result_tags_print.append(f'{result_dict[tag]} {tag}')
-
-    # sort tags
-    result_tags_out = []
-    sort_ndx = 0
-    if alpha_sort:
-        sort_ndx = 1
-
-    # sort by reverse by likelihood and normal for alpha, and format tag text as requested
-    unsorted_tags_in_theshold.sort(key=lambda y: y[sort_ndx], reverse=(not alpha_sort))
-    for weight, tag in unsorted_tags_in_theshold:
-        tag_outformat = tag
-        if use_spaces:
-            tag_outformat = tag_outformat.replace('_', ' ')
-        if use_escape:
-            tag_outformat = re.sub(re_special, r'\\\1', tag_outformat)
-        if include_ranks:
-            tag_outformat = f"({tag_outformat}:{weight:.3f})"
-
-        result_tags_out.append(tag_outformat)
-
-    print('\n'.join(sorted(result_tags_print, reverse=True)))
-
-    return ', '.join(result_tags_out)
+
+            probability_dict[tag] = probability
+
+        if alpha_sort:
+            tags = sorted(probability_dict)
+        else:
+            tags = [tag for tag, _ in sorted(probability_dict.items(), key=lambda x: -x[1])]
+
+        res = []
+
+        for tag in tags:
+            probability = probability_dict[tag]
+            tag_outformat = tag
+            if use_spaces:
+                tag_outformat = tag_outformat.replace('_', ' ')
+            if use_escape:
+                tag_outformat = re.sub(re_special, r'\\\1', tag_outformat)
+            if include_ranks:
+                tag_outformat = f"({tag_outformat}:{probability:.3f})"
+
+            res.append(tag_outformat)
+
+        return ", ".join(res)
+
+
+model = DeepDanbooru()

modules/shared.py

@@ -55,7 +55,7 @@ parser.add_argument("--ldsr-models-path", type=str, help="Path to directory with
 parser.add_argument("--clip-models-path", type=str, help="Path to directory with CLIP model file(s).", default=None)
 parser.add_argument("--xformers", action='store_true', help="enable xformers for cross attention layers")
 parser.add_argument("--force-enable-xformers", action='store_true', help="enable xformers for cross attention layers regardless of whether the checking code thinks you can run it; do not make bug reports if this fails to work")
-parser.add_argument("--deepdanbooru", action='store_true', help="enable deepdanbooru interrogator")
+parser.add_argument("--deepdanbooru", action='store_true', help="does not do anything")
 parser.add_argument("--opt-split-attention", action='store_true', help="force-enables Doggettx's cross-attention layer optimization. By default, it's on for torch cuda.")
 parser.add_argument("--opt-split-attention-invokeai", action='store_true', help="force-enables InvokeAI's cross-attention layer optimization. By default, it's on when cuda is unavailable.")
 parser.add_argument("--opt-split-attention-v1", action='store_true', help="enable older version of split attention optimization that does not consume all the VRAM it can find")

modules/textual_inversion/preprocess.py

@@ -6,12 +6,10 @@ import sys
 import tqdm
 import time
 
-from modules import shared, images
+from modules import shared, images, deepbooru
 from modules.paths import models_path
 from modules.shared import opts, cmd_opts
 from modules.textual_inversion import autocrop
-if cmd_opts.deepdanbooru:
-    import modules.deepbooru as deepbooru
 
 
 def preprocess(process_src, process_dst, process_width, process_height, preprocess_txt_action, process_flip, process_split, process_caption, process_caption_deepbooru=False, split_threshold=0.5, overlap_ratio=0.2, process_focal_crop=False, process_focal_crop_face_weight=0.9, process_focal_crop_entropy_weight=0.3, process_focal_crop_edges_weight=0.5, process_focal_crop_debug=False):
@@ -20,9 +18,7 @@ def preprocess(process_src, process_dst, process_width, process_height, preproce
             shared.interrogator.load()
 
         if process_caption_deepbooru:
-            db_opts = deepbooru.create_deepbooru_opts()
-            db_opts[deepbooru.OPT_INCLUDE_RANKS] = False
-            deepbooru.create_deepbooru_process(opts.interrogate_deepbooru_score_threshold, db_opts)
+            deepbooru.model.start()
 
         preprocess_work(process_src, process_dst, process_width, process_height, preprocess_txt_action, process_flip, process_split, process_caption, process_caption_deepbooru, split_threshold, overlap_ratio, process_focal_crop, process_focal_crop_face_weight, process_focal_crop_entropy_weight, process_focal_crop_edges_weight, process_focal_crop_debug)
 
@@ -32,7 +28,7 @@ def preprocess(process_src, process_dst, process_width, process_height, preproce
             shared.interrogator.send_blip_to_ram()
 
         if process_caption_deepbooru:
-            deepbooru.release_process()
+            deepbooru.model.stop()
 
 
 def listfiles(dirname):
@@ -58,7 +54,7 @@ def save_pic_with_caption(image, index, params: PreprocessParams, existing_capti
     if params.process_caption_deepbooru:
         if len(caption) > 0:
             caption += ", "
-        caption += deepbooru.get_tags_from_process(image)
+        caption += deepbooru.model.tag_multi(image)
 
     filename_part = params.src
     filename_part = os.path.splitext(filename_part)[0]

modules/ui.py

@@ -19,14 +19,11 @@ import numpy as np
 from PIL import Image, PngImagePlugin
 
 
-from modules import sd_hijack, sd_models, localization, script_callbacks, ui_extensions
+from modules import sd_hijack, sd_models, localization, script_callbacks, ui_extensions, deepbooru
 from modules.paths import script_path
 
 from modules.shared import opts, cmd_opts, restricted_opts
 
-if cmd_opts.deepdanbooru:
-    from modules.deepbooru import get_deepbooru_tags
-
 import modules.codeformer_model
 import modules.generation_parameters_copypaste as parameters_copypaste
 import modules.gfpgan_model
@@ -352,7 +349,7 @@ def interrogate(image):
 
 
 def interrogate_deepbooru(image):
-    prompt = get_deepbooru_tags(image)
+    prompt = deepbooru.model.tag(image)
     return gr_show(True) if prompt is None else prompt