From c7543d4940da672d970124ae8f2fec9de7bdc1da Mon Sep 17 00:00:00 2001
From: AUTOMATIC <16777216c@gmail.com>
Date: Sun, 2 Oct 2022 22:41:21 +0300
Subject: preprocessing for textual inversion added

---
 modules/textual_inversion/preprocess.py | 75 +++++++++++++++++++++++++++++++++
 1 file changed, 75 insertions(+)
 create mode 100644 modules/textual_inversion/preprocess.py

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
new file mode 100644
index 00000000..209e928f
--- /dev/null
+++ b/modules/textual_inversion/preprocess.py
@@ -0,0 +1,75 @@
+import os
+from PIL import Image, ImageOps
+import tqdm
+
+from modules import shared, images
+
+
+def preprocess(process_src, process_dst, process_flip, process_split, process_caption):
+    size = 512
+    src = os.path.abspath(process_src)
+    dst = os.path.abspath(process_dst)
+
+    assert src != dst, 'same directory specified as source and desitnation'
+
+    os.makedirs(dst, exist_ok=True)
+
+    files = os.listdir(src)
+
+    shared.state.textinfo = "Preprocessing..."
+    shared.state.job_count = len(files)
+
+    if process_caption:
+        shared.interrogator.load()
+
+    def save_pic_with_caption(image, index):
+        if process_caption:
+            caption = "-" + shared.interrogator.generate_caption(image)
+        else:
+            caption = ""
+
+        image.save(os.path.join(dst, f"{index:05}-{subindex[0]}{caption}.png"))
+        subindex[0] += 1
+
+    def save_pic(image, index):
+        save_pic_with_caption(image, index)
+
+        if process_flip:
+            save_pic_with_caption(ImageOps.mirror(image), index)
+
+    for index, imagefile in enumerate(tqdm.tqdm(files)):
+        subindex = [0]
+        filename = os.path.join(src, imagefile)
+        img = Image.open(filename).convert("RGB")
+
+        if shared.state.interrupted:
+            break
+
+        ratio = img.height / img.width
+        is_tall = ratio > 1.35
+        is_wide = ratio < 1 / 1.35
+
+        if process_split and is_tall:
+            img = img.resize((size, size * img.height // img.width))
+
+            top = img.crop((0, 0, size, size))
+            save_pic(top, index)
+
+            bot = img.crop((0, img.height - size, size, img.height))
+            save_pic(bot, index)
+        elif process_split and is_wide:
+            img = img.resize((size * img.width // img.height, size))
+
+            left = img.crop((0, 0, size, size))
+            save_pic(left, index)
+
+            right = img.crop((img.width - size, 0, img.width, size))
+            save_pic(right, index)
+        else:
+            img = images.resize_image(1, img, size, size)
+            save_pic(img, index)
+
+        shared.state.nextjob()
+
+    if process_caption:
+        shared.interrogator.send_blip_to_ram()
-- 
cgit v1.2.3


From 5ef0baf5eaec7f21a1666af424405cbee19f3764 Mon Sep 17 00:00:00 2001
From: AUTOMATIC <16777216c@gmail.com>
Date: Tue, 4 Oct 2022 08:52:11 +0300
Subject: add support for gelbooru tags in filenames for textual inversion

---
 modules/textual_inversion/dataset.py    | 7 +++++--
 modules/textual_inversion/preprocess.py | 4 +++-
 2 files changed, 8 insertions(+), 3 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py
index e8394ff6..7c44ea5b 100644
--- a/modules/textual_inversion/dataset.py
+++ b/modules/textual_inversion/dataset.py
@@ -9,6 +9,9 @@ from torchvision import transforms
 import random
 import tqdm
 from modules import devices
+import re
+
+re_tag = re.compile(r"[a-zA-Z][_\w\d()]+")
 
 
 class PersonalizedBase(Dataset):
@@ -38,8 +41,8 @@ class PersonalizedBase(Dataset):
             image = image.resize((self.width, self.height), PIL.Image.BICUBIC)
 
             filename = os.path.basename(path)
-            filename_tokens = os.path.splitext(filename)[0].replace('_', '-').replace(' ', '-').split('-')
-            filename_tokens = [token for token in filename_tokens if token.isalpha()]
+            filename_tokens = os.path.splitext(filename)[0]
+            filename_tokens = re_tag.findall(filename_tokens)
 
             npimage = np.array(image).astype(np.uint8)
             npimage = (npimage / 127.5 - 1.0).astype(np.float32)
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index 209e928f..f545a993 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -26,7 +26,9 @@ def preprocess(process_src, process_dst, process_flip, process_split, process_ca
         if process_caption:
             caption = "-" + shared.interrogator.generate_caption(image)
         else:
-            caption = ""
+            caption = filename
+            caption = os.path.splitext(caption)[0]
+            caption = os.path.basename(caption)
 
         image.save(os.path.join(dst, f"{index:05}-{subindex[0]}{caption}.png"))
         subindex[0] += 1
-- 
cgit v1.2.3


From 2499fb4e1910d31ff12c24110f161b20641b8835 Mon Sep 17 00:00:00 2001
From: Raphael Stoeckli <github@rabanti.ch>
Date: Wed, 5 Oct 2022 21:57:18 +0200
Subject: Add sanitizer for captions in Textual inversion

---
 modules/textual_inversion/preprocess.py | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index f545a993..4f3df4bd 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -1,5 +1,8 @@
+from cmath import log
 import os
 from PIL import Image, ImageOps
+import platform
+import sys
 import tqdm
 
 from modules import shared, images
@@ -25,6 +28,7 @@ def preprocess(process_src, process_dst, process_flip, process_split, process_ca
     def save_pic_with_caption(image, index):
         if process_caption:
             caption = "-" + shared.interrogator.generate_caption(image)
+            caption = sanitize_caption(os.path.join(dst, f"{index:05}-{subindex[0]}"), caption, ".png")
         else:
             caption = filename
             caption = os.path.splitext(caption)[0]
@@ -75,3 +79,27 @@ def preprocess(process_src, process_dst, process_flip, process_split, process_ca
 
     if process_caption:
         shared.interrogator.send_blip_to_ram()
+
+def sanitize_caption(base_path, original_caption, suffix):
+    operating_system = platform.system().lower()
+    if (operating_system == "windows"):
+        invalid_path_characters = "\\/:*?\"<>|"
+        max_path_length = 259
+    else:
+        invalid_path_characters = "/" #linux/macos
+        max_path_length = 1023
+    caption = original_caption
+    for invalid_character in invalid_path_characters:
+        caption = caption.replace(invalid_character, "")
+    fixed_path_length = len(base_path) + len(suffix) 
+    if fixed_path_length + len(caption) <= max_path_length:
+        return caption
+    caption_tokens = caption.split()
+    new_caption = ""
+    for token in caption_tokens:
+        last_caption = new_caption
+        new_caption = new_caption + token + " "
+        if (len(new_caption) + fixed_path_length - 1  > max_path_length):
+            break
+    print(f"\nPath will be too long. Truncated caption: {original_caption}\nto: {last_caption}", file=sys.stderr)
+    return last_caption.strip()
-- 
cgit v1.2.3


From 4288e53fc2ea25fa49715bf5b7f14603553c9e38 Mon Sep 17 00:00:00 2001
From: Raphael Stoeckli <github@rabanti.ch>
Date: Wed, 5 Oct 2022 23:11:32 +0200
Subject: removed unused import, fixed typo

---
 modules/textual_inversion/preprocess.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index 4f3df4bd..f1c002a2 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -1,4 +1,3 @@
-from cmath import log
 import os
 from PIL import Image, ImageOps
 import platform
@@ -13,7 +12,7 @@ def preprocess(process_src, process_dst, process_flip, process_split, process_ca
     src = os.path.abspath(process_src)
     dst = os.path.abspath(process_dst)
 
-    assert src != dst, 'same directory specified as source and desitnation'
+    assert src != dst, 'same directory specified as source and destination'
 
     os.makedirs(dst, exist_ok=True)
 
-- 
cgit v1.2.3


From 1f92336be768d235c18a82acb2195b7135101ae7 Mon Sep 17 00:00:00 2001
From: JC_Array <justin318@att.net>
Date: Sun, 9 Oct 2022 23:58:18 -0500
Subject: refactored the deepbooru module to improve speed on running multiple
 interogations in a row.  Added the option to generate deepbooru tags for
 textual inversion preproccessing.

---
 modules/deepbooru.py                    | 84 +++++++++++++++++++++++++--------
 modules/textual_inversion/preprocess.py | 22 ++++++++-
 modules/ui.py                           | 52 ++++++++++++++------
 3 files changed, 122 insertions(+), 36 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/deepbooru.py b/modules/deepbooru.py
index 7e3c0618..cee4a3b4 100644
--- a/modules/deepbooru.py
+++ b/modules/deepbooru.py
@@ -1,21 +1,74 @@
 import os.path
 from concurrent.futures import ProcessPoolExecutor
-from multiprocessing import get_context
+import multiprocessing
 
 
-def _load_tf_and_return_tags(pil_image, threshold):
+def get_deepbooru_tags(pil_image, threshold=0.5):
+    """
+    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
+    create_deepbooru_process(threshold)
+    shared.deepbooru_process_return["value"] = -1
+    shared.deepbooru_process_queue.put(pil_image)
+    while shared.deepbooru_process_return["value"] == -1:
+        time.sleep(0.2)
+    release_process()
+    return ret
+
+
+def deepbooru_process(queue, deepbooru_process_return, threshold):
+    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)
+
+
+def create_deepbooru_process(threshold=0.5):
+    """
+    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
+    shared.deepbooru_process_manager = multiprocessing.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 = multiprocessing.Process(target=deepbooru_process, args=(shared.deepbooru_process_queue, shared.deepbooru_process_return, threshold))
+    shared.deepbooru_process.start()
+
+
+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)
+        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"))
@@ -24,7 +77,13 @@ def _load_tf_and_return_tags(pil_image, threshold):
     model = dd.project.load_model_from_project(
         model_path, compile_model=True
     )
+    return model, tags
 
+
+def get_deepbooru_tags_from_model(model, tags, pil_image, threshold=0.5):
+    import deepdanbooru as dd
+    import tensorflow as tf
+    import numpy as np
     width = model.input_shape[2]
     height = model.input_shape[1]
     image = np.array(pil_image)
@@ -57,17 +116,4 @@ def _load_tf_and_return_tags(pil_image, threshold):
 
     print('\n'.join(sorted(result_tags_print, reverse=True)))
 
-    return ', '.join(result_tags_out).replace('_', ' ').replace(':', ' ')
-
-
-def subprocess_init_no_cuda():
-    import os
-    os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
-
-
-def get_deepbooru_tags(pil_image, threshold=0.5):
-    context = get_context('spawn')
-    with ProcessPoolExecutor(initializer=subprocess_init_no_cuda, mp_context=context) as executor:
-        f = executor.submit(_load_tf_and_return_tags, pil_image, threshold, )
-        ret = f.result()  # will rethrow any exceptions
-    return ret
\ No newline at end of file
+    return ', '.join(result_tags_out).replace('_', ' ').replace(':', ' ')
\ No newline at end of file
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index f1c002a2..9f63c9a4 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -3,11 +3,14 @@ from PIL import Image, ImageOps
 import platform
 import sys
 import tqdm
+import time
 
 from modules import shared, images
+from modules.shared import opts, cmd_opts
+if cmd_opts.deepdanbooru:
+    import modules.deepbooru as deepbooru
 
-
-def preprocess(process_src, process_dst, process_flip, process_split, process_caption):
+def preprocess(process_src, process_dst, process_flip, process_split, process_caption, process_caption_deepbooru=False):
     size = 512
     src = os.path.abspath(process_src)
     dst = os.path.abspath(process_dst)
@@ -24,10 +27,21 @@ def preprocess(process_src, process_dst, process_flip, process_split, process_ca
     if process_caption:
         shared.interrogator.load()
 
+    if process_caption_deepbooru:
+        deepbooru.create_deepbooru_process()
+
     def save_pic_with_caption(image, index):
         if process_caption:
             caption = "-" + shared.interrogator.generate_caption(image)
             caption = sanitize_caption(os.path.join(dst, f"{index:05}-{subindex[0]}"), caption, ".png")
+        elif process_caption_deepbooru:
+            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"]
+            caption = sanitize_caption(os.path.join(dst, f"{index:05}-{subindex[0]}"), caption, ".png")
+            shared.deepbooru_process_return["value"] = -1
         else:
             caption = filename
             caption = os.path.splitext(caption)[0]
@@ -79,6 +93,10 @@ def preprocess(process_src, process_dst, process_flip, process_split, process_ca
     if process_caption:
         shared.interrogator.send_blip_to_ram()
 
+    if process_caption_deepbooru:
+        deepbooru.release_process()
+
+
 def sanitize_caption(base_path, original_caption, suffix):
     operating_system = platform.system().lower()
     if (operating_system == "windows"):
diff --git a/modules/ui.py b/modules/ui.py
index 2231a8ed..179e3a83 100644
--- a/modules/ui.py
+++ b/modules/ui.py
@@ -1034,6 +1034,9 @@ def create_ui(wrap_gradio_gpu_call):
                         process_flip = gr.Checkbox(label='Create flipped copies')
                         process_split = gr.Checkbox(label='Split oversized images into two')
                         process_caption = gr.Checkbox(label='Use BLIP caption as filename')
+                        if cmd_opts.deepdanbooru:
+                            process_caption_deepbooru = gr.Checkbox(label='Use deepbooru caption as filename')
+
 
                     with gr.Row():
                         with gr.Column(scale=3):
@@ -1086,21 +1089,40 @@ def create_ui(wrap_gradio_gpu_call):
             ]
         )
 
-        run_preprocess.click(
-            fn=wrap_gradio_gpu_call(modules.textual_inversion.ui.preprocess, extra_outputs=[gr.update()]),
-            _js="start_training_textual_inversion",
-            inputs=[
-                process_src,
-                process_dst,
-                process_flip,
-                process_split,
-                process_caption,
-            ],
-            outputs=[
-                ti_output,
-                ti_outcome,
-            ],
-        )
+        if cmd_opts.deepdanbooru:
+            # if process_caption_deepbooru is None, it will cause an error, as a result only include it if it is enabled
+            run_preprocess.click(
+                fn=wrap_gradio_gpu_call(modules.textual_inversion.ui.preprocess, extra_outputs=[gr.update()]),
+                _js="start_training_textual_inversion",
+                inputs=[
+                    process_src,
+                    process_dst,
+                    process_flip,
+                    process_split,
+                    process_caption,
+                    process_caption_deepbooru,
+                ],
+                outputs=[
+                    ti_output,
+                    ti_outcome,
+                ],
+            )
+        else:
+            run_preprocess.click(
+                fn=wrap_gradio_gpu_call(modules.textual_inversion.ui.preprocess, extra_outputs=[gr.update()]),
+                _js="start_training_textual_inversion",
+                inputs=[
+                    process_src,
+                    process_dst,
+                    process_flip,
+                    process_split,
+                    process_caption,
+                ],
+                outputs=[
+                    ti_output,
+                    ti_outcome,
+                ],
+            )
 
         train_embedding.click(
             fn=wrap_gradio_gpu_call(modules.textual_inversion.ui.train_embedding, extra_outputs=[gr.update()]),
-- 
cgit v1.2.3


From 3110f895b2718a3a25aae419fdf5c87c177ec9f4 Mon Sep 17 00:00:00 2001
From: alg-wiki <alg.4chan@gmail.com>
Date: Mon, 10 Oct 2022 17:07:46 +0900
Subject: Textual Inversion: Added custom training image size and number of
 repeats per input image in a single epoch

---
 modules/textual_inversion/dataset.py           |  6 +++---
 modules/textual_inversion/preprocess.py        |  4 ++--
 modules/textual_inversion/textual_inversion.py | 15 ++++++++++++---
 modules/ui.py                                  |  8 +++++++-
 4 files changed, 24 insertions(+), 9 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py
index 7c44ea5b..acc4ce59 100644
--- a/modules/textual_inversion/dataset.py
+++ b/modules/textual_inversion/dataset.py
@@ -15,13 +15,13 @@ re_tag = re.compile(r"[a-zA-Z][_\w\d()]+")
 
 
 class PersonalizedBase(Dataset):
-    def __init__(self, data_root, size=None, repeats=100, flip_p=0.5, placeholder_token="*", width=512, height=512, model=None, device=None, template_file=None):
+    def __init__(self, data_root, size, repeats, flip_p=0.5, placeholder_token="*", model=None, device=None, template_file=None):
 
         self.placeholder_token = placeholder_token
 
         self.size = size
-        self.width = width
-        self.height = height
+        self.width = size
+        self.height = size
         self.flip = transforms.RandomHorizontalFlip(p=flip_p)
 
         self.dataset = []
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index f1c002a2..b3de6fd7 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -7,8 +7,8 @@ import tqdm
 from modules import shared, images
 
 
-def preprocess(process_src, process_dst, process_flip, process_split, process_caption):
-    size = 512
+def preprocess(process_src, process_dst, process_size, process_flip, process_split, process_caption):
+    size = process_size
     src = os.path.abspath(process_src)
     dst = os.path.abspath(process_dst)
 
diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py
index cd9f3498..e34dc2e8 100644
--- a/modules/textual_inversion/textual_inversion.py
+++ b/modules/textual_inversion/textual_inversion.py
@@ -6,6 +6,7 @@ import torch
 import tqdm
 import html
 import datetime
+import math
 
 
 from modules import shared, devices, sd_hijack, processing, sd_models
@@ -156,7 +157,7 @@ def create_embedding(name, num_vectors_per_token, init_text='*'):
     return fn
 
 
-def train_embedding(embedding_name, learn_rate, data_root, log_directory, steps, create_image_every, save_embedding_every, template_file):
+def train_embedding(embedding_name, learn_rate, data_root, log_directory, training_size, steps, num_repeats, create_image_every, save_embedding_every, template_file):
     assert embedding_name, 'embedding not selected'
 
     shared.state.textinfo = "Initializing textual inversion training..."
@@ -182,7 +183,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, steps,
 
     shared.state.textinfo = f"Preparing dataset from {html.escape(data_root)}..."
     with torch.autocast("cuda"):
-        ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, size=512, placeholder_token=embedding_name, model=shared.sd_model, device=devices.device, template_file=template_file)
+        ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, size=training_size, repeats=num_repeats, placeholder_token=embedding_name, model=shared.sd_model, device=devices.device, template_file=template_file)
 
     hijack = sd_hijack.model_hijack
 
@@ -200,6 +201,9 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, steps,
     if ititial_step > steps:
         return embedding, filename
 
+    tr_img_len = len([os.path.join(data_root, file_path) for file_path in os.listdir(data_root)])
+    epoch_len = (tr_img_len * num_repeats) + tr_img_len
+
     pbar = tqdm.tqdm(enumerate(ds), total=steps-ititial_step)
     for i, (x, text) in pbar:
         embedding.step = i + ititial_step
@@ -223,7 +227,10 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, steps,
             loss.backward()
             optimizer.step()
 
-        pbar.set_description(f"loss: {losses.mean():.7f}")
+        epoch_num = math.floor(embedding.step / epoch_len)
+        epoch_step = embedding.step - (epoch_num * epoch_len)
+
+        pbar.set_description(f"[Epoch {epoch_num}: {epoch_step}/{epoch_len}]loss: {losses.mean():.7f}")
 
         if embedding.step > 0 and embedding_dir is not None and embedding.step % save_embedding_every == 0:
             last_saved_file = os.path.join(embedding_dir, f'{embedding_name}-{embedding.step}.pt')
@@ -236,6 +243,8 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, steps,
                 sd_model=shared.sd_model,
                 prompt=text,
                 steps=20,
+                height=training_size,
+                width=training_size,
                 do_not_save_grid=True,
                 do_not_save_samples=True,
             )
diff --git a/modules/ui.py b/modules/ui.py
index 2231a8ed..f821fd8d 100644
--- a/modules/ui.py
+++ b/modules/ui.py
@@ -1029,6 +1029,7 @@ def create_ui(wrap_gradio_gpu_call):
 
                     process_src = gr.Textbox(label='Source directory')
                     process_dst = gr.Textbox(label='Destination directory')
+                    process_size = gr.Slider(minimum=64, maximum=2048, step=64, label="Size (width and height)", value=512)
 
                     with gr.Row():
                         process_flip = gr.Checkbox(label='Create flipped copies')
@@ -1043,13 +1044,15 @@ def create_ui(wrap_gradio_gpu_call):
                             run_preprocess = gr.Button(value="Preprocess", variant='primary')
 
                 with gr.Group():
-                    gr.HTML(value="<p style='margin-bottom: 0.7em'>Train an embedding; must specify a directory with a set of 512x512 images</p>")
+                    gr.HTML(value="<p style='margin-bottom: 0.7em'>Train an embedding; must specify a directory with a set of 1:1 ratio images</p>")
                     train_embedding_name = gr.Dropdown(label='Embedding', choices=sorted(sd_hijack.model_hijack.embedding_db.word_embeddings.keys()))
                     learn_rate = gr.Number(label='Learning rate', value=5.0e-03)
                     dataset_directory = gr.Textbox(label='Dataset directory', placeholder="Path to directory with input images")
                     log_directory = gr.Textbox(label='Log directory', placeholder="Path to directory where to write outputs", value="textual_inversion")
                     template_file = gr.Textbox(label='Prompt template file', value=os.path.join(script_path, "textual_inversion_templates", "style_filewords.txt"))
+                    training_size = gr.Slider(minimum=64, maximum=2048, step=64, label="Size (width and height)", value=512)
                     steps = gr.Number(label='Max steps', value=100000, precision=0)
+                    num_repeats = gr.Number(label='Number of repeats for a single input image per epoch', value=100, precision=0)
                     create_image_every = gr.Number(label='Save an image to log directory every N steps, 0 to disable', value=500, precision=0)
                     save_embedding_every = gr.Number(label='Save a copy of embedding to log directory every N steps, 0 to disable', value=500, precision=0)
 
@@ -1092,6 +1095,7 @@ def create_ui(wrap_gradio_gpu_call):
             inputs=[
                 process_src,
                 process_dst,
+                process_size,
                 process_flip,
                 process_split,
                 process_caption,
@@ -1110,7 +1114,9 @@ def create_ui(wrap_gradio_gpu_call):
                 learn_rate,
                 dataset_directory,
                 log_directory,
+                training_size,
                 steps,
+                num_repeats,
                 create_image_every,
                 save_embedding_every,
                 template_file,
-- 
cgit v1.2.3


From 04c745ea4f81518999927fee5f78500560c25e29 Mon Sep 17 00:00:00 2001
From: alg-wiki <alg.4chan@gmail.com>
Date: Mon, 10 Oct 2022 22:35:35 +0900
Subject: Custom Width and Height

---
 modules/textual_inversion/dataset.py           |  7 +++----
 modules/textual_inversion/preprocess.py        | 19 ++++++++++---------
 modules/textual_inversion/textual_inversion.py | 11 +++++------
 modules/ui.py                                  | 12 ++++++++----
 4 files changed, 26 insertions(+), 23 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py
index acc4ce59..bcf772d2 100644
--- a/modules/textual_inversion/dataset.py
+++ b/modules/textual_inversion/dataset.py
@@ -15,13 +15,12 @@ re_tag = re.compile(r"[a-zA-Z][_\w\d()]+")
 
 
 class PersonalizedBase(Dataset):
-    def __init__(self, data_root, size, repeats, flip_p=0.5, placeholder_token="*", model=None, device=None, template_file=None):
+    def __init__(self, data_root, width, height, repeats, flip_p=0.5, placeholder_token="*", model=None, device=None, template_file=None):
 
         self.placeholder_token = placeholder_token
 
-        self.size = size
-        self.width = size
-        self.height = size
+        self.width = width
+        self.height = height
         self.flip = transforms.RandomHorizontalFlip(p=flip_p)
 
         self.dataset = []
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index b3de6fd7..d7efdef2 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -7,8 +7,9 @@ import tqdm
 from modules import shared, images
 
 
-def preprocess(process_src, process_dst, process_size, process_flip, process_split, process_caption):
-    size = process_size
+def preprocess(process_src, process_dst, process_width, process_height, process_flip, process_split, process_caption):
+    width = process_width
+    height = process_height
     src = os.path.abspath(process_src)
     dst = os.path.abspath(process_dst)
 
@@ -55,23 +56,23 @@ def preprocess(process_src, process_dst, process_size, process_flip, process_spl
         is_wide = ratio < 1 / 1.35
 
         if process_split and is_tall:
-            img = img.resize((size, size * img.height // img.width))
+            img = img.resize((width, height * img.height // img.width))
 
-            top = img.crop((0, 0, size, size))
+            top = img.crop((0, 0, width, height))
             save_pic(top, index)
 
-            bot = img.crop((0, img.height - size, size, img.height))
+            bot = img.crop((0, img.height - height, width, img.height))
             save_pic(bot, index)
         elif process_split and is_wide:
-            img = img.resize((size * img.width // img.height, size))
+            img = img.resize((width * img.width // img.height, height))
 
-            left = img.crop((0, 0, size, size))
+            left = img.crop((0, 0, width, height))
             save_pic(left, index)
 
-            right = img.crop((img.width - size, 0, img.width, size))
+            right = img.crop((img.width - width, 0, img.width, height))
             save_pic(right, index)
         else:
-            img = images.resize_image(1, img, size, size)
+            img = images.resize_image(1, img, width, height)
             save_pic(img, index)
 
         shared.state.nextjob()
diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py
index 769682ea..5965c5a0 100644
--- a/modules/textual_inversion/textual_inversion.py
+++ b/modules/textual_inversion/textual_inversion.py
@@ -6,7 +6,6 @@ import torch
 import tqdm
 import html
 import datetime
-import math
 
 
 from modules import shared, devices, sd_hijack, processing, sd_models
@@ -157,7 +156,7 @@ def create_embedding(name, num_vectors_per_token, init_text='*'):
     return fn
 
 
-def train_embedding(embedding_name, learn_rate, data_root, log_directory, training_size, steps, num_repeats, create_image_every, save_embedding_every, template_file):
+def train_embedding(embedding_name, learn_rate, data_root, log_directory, training_width, training_height, steps, num_repeats, create_image_every, save_embedding_every, template_file):
     assert embedding_name, 'embedding not selected'
 
     shared.state.textinfo = "Initializing textual inversion training..."
@@ -183,7 +182,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
 
     shared.state.textinfo = f"Preparing dataset from {html.escape(data_root)}..."
     with torch.autocast("cuda"):
-        ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, size=training_size, repeats=num_repeats, placeholder_token=embedding_name, model=shared.sd_model, device=devices.device, template_file=template_file)
+        ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=training_width, height=training_height, repeats=num_repeats, placeholder_token=embedding_name, model=shared.sd_model, device=devices.device, template_file=template_file)
 
     hijack = sd_hijack.model_hijack
 
@@ -227,7 +226,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
             loss.backward()
             optimizer.step()
 
-        epoch_num = math.floor(embedding.step / epoch_len)
+        epoch_num = embedding.step // epoch_len
         epoch_step = embedding.step - (epoch_num * epoch_len) + 1
 
         pbar.set_description(f"[Epoch {epoch_num}: {epoch_step}/{epoch_len}]loss: {losses.mean():.7f}")
@@ -243,8 +242,8 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
                 sd_model=shared.sd_model,
                 prompt=text,
                 steps=20,
-                height=training_size,
-                width=training_size,
+				height=training_height,
+				width=training_width,
                 do_not_save_grid=True,
                 do_not_save_samples=True,
             )
diff --git a/modules/ui.py b/modules/ui.py
index f821fd8d..8c06ad7c 100644
--- a/modules/ui.py
+++ b/modules/ui.py
@@ -1029,7 +1029,8 @@ def create_ui(wrap_gradio_gpu_call):
 
                     process_src = gr.Textbox(label='Source directory')
                     process_dst = gr.Textbox(label='Destination directory')
-                    process_size = gr.Slider(minimum=64, maximum=2048, step=64, label="Size (width and height)", value=512)
+                    process_width = gr.Slider(minimum=64, maximum=2048, step=64, label="Width", value=512)
+                    process_height = gr.Slider(minimum=64, maximum=2048, step=64, label="Height", value=512)
 
                     with gr.Row():
                         process_flip = gr.Checkbox(label='Create flipped copies')
@@ -1050,7 +1051,8 @@ def create_ui(wrap_gradio_gpu_call):
                     dataset_directory = gr.Textbox(label='Dataset directory', placeholder="Path to directory with input images")
                     log_directory = gr.Textbox(label='Log directory', placeholder="Path to directory where to write outputs", value="textual_inversion")
                     template_file = gr.Textbox(label='Prompt template file', value=os.path.join(script_path, "textual_inversion_templates", "style_filewords.txt"))
-                    training_size = gr.Slider(minimum=64, maximum=2048, step=64, label="Size (width and height)", value=512)
+                    training_width = gr.Slider(minimum=64, maximum=2048, step=64, label="Width", value=512)
+                    training_height = gr.Slider(minimum=64, maximum=2048, step=64, label="Height", value=512)
                     steps = gr.Number(label='Max steps', value=100000, precision=0)
                     num_repeats = gr.Number(label='Number of repeats for a single input image per epoch', value=100, precision=0)
                     create_image_every = gr.Number(label='Save an image to log directory every N steps, 0 to disable', value=500, precision=0)
@@ -1095,7 +1097,8 @@ def create_ui(wrap_gradio_gpu_call):
             inputs=[
                 process_src,
                 process_dst,
-                process_size,
+                process_width,
+                process_height,
                 process_flip,
                 process_split,
                 process_caption,
@@ -1114,7 +1117,8 @@ def create_ui(wrap_gradio_gpu_call):
                 learn_rate,
                 dataset_directory,
                 log_directory,
-                training_size,
+                training_width,
+                training_height,
                 steps,
                 num_repeats,
                 create_image_every,
-- 
cgit v1.2.3


From ea00c1624bbb0dcb5be07f59c9509061baddf5b1 Mon Sep 17 00:00:00 2001
From: alg-wiki <alg.4chan@gmail.com>
Date: Mon, 10 Oct 2022 17:07:46 +0900
Subject: Textual Inversion: Added custom training image size and number of
 repeats per input image in a single epoch

---
 modules/textual_inversion/dataset.py           |  6 +++---
 modules/textual_inversion/preprocess.py        |  4 ++--
 modules/textual_inversion/textual_inversion.py | 15 ++++++++++++---
 modules/ui.py                                  |  8 +++++++-
 4 files changed, 24 insertions(+), 9 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py
index 7c44ea5b..acc4ce59 100644
--- a/modules/textual_inversion/dataset.py
+++ b/modules/textual_inversion/dataset.py
@@ -15,13 +15,13 @@ re_tag = re.compile(r"[a-zA-Z][_\w\d()]+")
 
 
 class PersonalizedBase(Dataset):
-    def __init__(self, data_root, size=None, repeats=100, flip_p=0.5, placeholder_token="*", width=512, height=512, model=None, device=None, template_file=None):
+    def __init__(self, data_root, size, repeats, flip_p=0.5, placeholder_token="*", model=None, device=None, template_file=None):
 
         self.placeholder_token = placeholder_token
 
         self.size = size
-        self.width = width
-        self.height = height
+        self.width = size
+        self.height = size
         self.flip = transforms.RandomHorizontalFlip(p=flip_p)
 
         self.dataset = []
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index f1c002a2..b3de6fd7 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -7,8 +7,8 @@ import tqdm
 from modules import shared, images
 
 
-def preprocess(process_src, process_dst, process_flip, process_split, process_caption):
-    size = 512
+def preprocess(process_src, process_dst, process_size, process_flip, process_split, process_caption):
+    size = process_size
     src = os.path.abspath(process_src)
     dst = os.path.abspath(process_dst)
 
diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py
index cd9f3498..e34dc2e8 100644
--- a/modules/textual_inversion/textual_inversion.py
+++ b/modules/textual_inversion/textual_inversion.py
@@ -6,6 +6,7 @@ import torch
 import tqdm
 import html
 import datetime
+import math
 
 
 from modules import shared, devices, sd_hijack, processing, sd_models
@@ -156,7 +157,7 @@ def create_embedding(name, num_vectors_per_token, init_text='*'):
     return fn
 
 
-def train_embedding(embedding_name, learn_rate, data_root, log_directory, steps, create_image_every, save_embedding_every, template_file):
+def train_embedding(embedding_name, learn_rate, data_root, log_directory, training_size, steps, num_repeats, create_image_every, save_embedding_every, template_file):
     assert embedding_name, 'embedding not selected'
 
     shared.state.textinfo = "Initializing textual inversion training..."
@@ -182,7 +183,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, steps,
 
     shared.state.textinfo = f"Preparing dataset from {html.escape(data_root)}..."
     with torch.autocast("cuda"):
-        ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, size=512, placeholder_token=embedding_name, model=shared.sd_model, device=devices.device, template_file=template_file)
+        ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, size=training_size, repeats=num_repeats, placeholder_token=embedding_name, model=shared.sd_model, device=devices.device, template_file=template_file)
 
     hijack = sd_hijack.model_hijack
 
@@ -200,6 +201,9 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, steps,
     if ititial_step > steps:
         return embedding, filename
 
+    tr_img_len = len([os.path.join(data_root, file_path) for file_path in os.listdir(data_root)])
+    epoch_len = (tr_img_len * num_repeats) + tr_img_len
+
     pbar = tqdm.tqdm(enumerate(ds), total=steps-ititial_step)
     for i, (x, text) in pbar:
         embedding.step = i + ititial_step
@@ -223,7 +227,10 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, steps,
             loss.backward()
             optimizer.step()
 
-        pbar.set_description(f"loss: {losses.mean():.7f}")
+        epoch_num = math.floor(embedding.step / epoch_len)
+        epoch_step = embedding.step - (epoch_num * epoch_len)
+
+        pbar.set_description(f"[Epoch {epoch_num}: {epoch_step}/{epoch_len}]loss: {losses.mean():.7f}")
 
         if embedding.step > 0 and embedding_dir is not None and embedding.step % save_embedding_every == 0:
             last_saved_file = os.path.join(embedding_dir, f'{embedding_name}-{embedding.step}.pt')
@@ -236,6 +243,8 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, steps,
                 sd_model=shared.sd_model,
                 prompt=text,
                 steps=20,
+                height=training_size,
+                width=training_size,
                 do_not_save_grid=True,
                 do_not_save_samples=True,
             )
diff --git a/modules/ui.py b/modules/ui.py
index 2231a8ed..f821fd8d 100644
--- a/modules/ui.py
+++ b/modules/ui.py
@@ -1029,6 +1029,7 @@ def create_ui(wrap_gradio_gpu_call):
 
                     process_src = gr.Textbox(label='Source directory')
                     process_dst = gr.Textbox(label='Destination directory')
+                    process_size = gr.Slider(minimum=64, maximum=2048, step=64, label="Size (width and height)", value=512)
 
                     with gr.Row():
                         process_flip = gr.Checkbox(label='Create flipped copies')
@@ -1043,13 +1044,15 @@ def create_ui(wrap_gradio_gpu_call):
                             run_preprocess = gr.Button(value="Preprocess", variant='primary')
 
                 with gr.Group():
-                    gr.HTML(value="<p style='margin-bottom: 0.7em'>Train an embedding; must specify a directory with a set of 512x512 images</p>")
+                    gr.HTML(value="<p style='margin-bottom: 0.7em'>Train an embedding; must specify a directory with a set of 1:1 ratio images</p>")
                     train_embedding_name = gr.Dropdown(label='Embedding', choices=sorted(sd_hijack.model_hijack.embedding_db.word_embeddings.keys()))
                     learn_rate = gr.Number(label='Learning rate', value=5.0e-03)
                     dataset_directory = gr.Textbox(label='Dataset directory', placeholder="Path to directory with input images")
                     log_directory = gr.Textbox(label='Log directory', placeholder="Path to directory where to write outputs", value="textual_inversion")
                     template_file = gr.Textbox(label='Prompt template file', value=os.path.join(script_path, "textual_inversion_templates", "style_filewords.txt"))
+                    training_size = gr.Slider(minimum=64, maximum=2048, step=64, label="Size (width and height)", value=512)
                     steps = gr.Number(label='Max steps', value=100000, precision=0)
+                    num_repeats = gr.Number(label='Number of repeats for a single input image per epoch', value=100, precision=0)
                     create_image_every = gr.Number(label='Save an image to log directory every N steps, 0 to disable', value=500, precision=0)
                     save_embedding_every = gr.Number(label='Save a copy of embedding to log directory every N steps, 0 to disable', value=500, precision=0)
 
@@ -1092,6 +1095,7 @@ def create_ui(wrap_gradio_gpu_call):
             inputs=[
                 process_src,
                 process_dst,
+                process_size,
                 process_flip,
                 process_split,
                 process_caption,
@@ -1110,7 +1114,9 @@ def create_ui(wrap_gradio_gpu_call):
                 learn_rate,
                 dataset_directory,
                 log_directory,
+                training_size,
                 steps,
+                num_repeats,
                 create_image_every,
                 save_embedding_every,
                 template_file,
-- 
cgit v1.2.3


From 7a20f914eddfdf09c0ccced157ec108205bc3d0f Mon Sep 17 00:00:00 2001
From: alg-wiki <alg.4chan@gmail.com>
Date: Mon, 10 Oct 2022 22:35:35 +0900
Subject: Custom Width and Height

---
 modules/textual_inversion/dataset.py           |  7 +++----
 modules/textual_inversion/preprocess.py        | 19 ++++++++++---------
 modules/textual_inversion/textual_inversion.py | 11 +++++------
 modules/ui.py                                  | 12 ++++++++----
 4 files changed, 26 insertions(+), 23 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py
index acc4ce59..bcf772d2 100644
--- a/modules/textual_inversion/dataset.py
+++ b/modules/textual_inversion/dataset.py
@@ -15,13 +15,12 @@ re_tag = re.compile(r"[a-zA-Z][_\w\d()]+")
 
 
 class PersonalizedBase(Dataset):
-    def __init__(self, data_root, size, repeats, flip_p=0.5, placeholder_token="*", model=None, device=None, template_file=None):
+    def __init__(self, data_root, width, height, repeats, flip_p=0.5, placeholder_token="*", model=None, device=None, template_file=None):
 
         self.placeholder_token = placeholder_token
 
-        self.size = size
-        self.width = size
-        self.height = size
+        self.width = width
+        self.height = height
         self.flip = transforms.RandomHorizontalFlip(p=flip_p)
 
         self.dataset = []
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index b3de6fd7..d7efdef2 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -7,8 +7,9 @@ import tqdm
 from modules import shared, images
 
 
-def preprocess(process_src, process_dst, process_size, process_flip, process_split, process_caption):
-    size = process_size
+def preprocess(process_src, process_dst, process_width, process_height, process_flip, process_split, process_caption):
+    width = process_width
+    height = process_height
     src = os.path.abspath(process_src)
     dst = os.path.abspath(process_dst)
 
@@ -55,23 +56,23 @@ def preprocess(process_src, process_dst, process_size, process_flip, process_spl
         is_wide = ratio < 1 / 1.35
 
         if process_split and is_tall:
-            img = img.resize((size, size * img.height // img.width))
+            img = img.resize((width, height * img.height // img.width))
 
-            top = img.crop((0, 0, size, size))
+            top = img.crop((0, 0, width, height))
             save_pic(top, index)
 
-            bot = img.crop((0, img.height - size, size, img.height))
+            bot = img.crop((0, img.height - height, width, img.height))
             save_pic(bot, index)
         elif process_split and is_wide:
-            img = img.resize((size * img.width // img.height, size))
+            img = img.resize((width * img.width // img.height, height))
 
-            left = img.crop((0, 0, size, size))
+            left = img.crop((0, 0, width, height))
             save_pic(left, index)
 
-            right = img.crop((img.width - size, 0, img.width, size))
+            right = img.crop((img.width - width, 0, img.width, height))
             save_pic(right, index)
         else:
-            img = images.resize_image(1, img, size, size)
+            img = images.resize_image(1, img, width, height)
             save_pic(img, index)
 
         shared.state.nextjob()
diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py
index 769682ea..5965c5a0 100644
--- a/modules/textual_inversion/textual_inversion.py
+++ b/modules/textual_inversion/textual_inversion.py
@@ -6,7 +6,6 @@ import torch
 import tqdm
 import html
 import datetime
-import math
 
 
 from modules import shared, devices, sd_hijack, processing, sd_models
@@ -157,7 +156,7 @@ def create_embedding(name, num_vectors_per_token, init_text='*'):
     return fn
 
 
-def train_embedding(embedding_name, learn_rate, data_root, log_directory, training_size, steps, num_repeats, create_image_every, save_embedding_every, template_file):
+def train_embedding(embedding_name, learn_rate, data_root, log_directory, training_width, training_height, steps, num_repeats, create_image_every, save_embedding_every, template_file):
     assert embedding_name, 'embedding not selected'
 
     shared.state.textinfo = "Initializing textual inversion training..."
@@ -183,7 +182,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
 
     shared.state.textinfo = f"Preparing dataset from {html.escape(data_root)}..."
     with torch.autocast("cuda"):
-        ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, size=training_size, repeats=num_repeats, placeholder_token=embedding_name, model=shared.sd_model, device=devices.device, template_file=template_file)
+        ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=training_width, height=training_height, repeats=num_repeats, placeholder_token=embedding_name, model=shared.sd_model, device=devices.device, template_file=template_file)
 
     hijack = sd_hijack.model_hijack
 
@@ -227,7 +226,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
             loss.backward()
             optimizer.step()
 
-        epoch_num = math.floor(embedding.step / epoch_len)
+        epoch_num = embedding.step // epoch_len
         epoch_step = embedding.step - (epoch_num * epoch_len) + 1
 
         pbar.set_description(f"[Epoch {epoch_num}: {epoch_step}/{epoch_len}]loss: {losses.mean():.7f}")
@@ -243,8 +242,8 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
                 sd_model=shared.sd_model,
                 prompt=text,
                 steps=20,
-                height=training_size,
-                width=training_size,
+				height=training_height,
+				width=training_width,
                 do_not_save_grid=True,
                 do_not_save_samples=True,
             )
diff --git a/modules/ui.py b/modules/ui.py
index f821fd8d..8c06ad7c 100644
--- a/modules/ui.py
+++ b/modules/ui.py
@@ -1029,7 +1029,8 @@ def create_ui(wrap_gradio_gpu_call):
 
                     process_src = gr.Textbox(label='Source directory')
                     process_dst = gr.Textbox(label='Destination directory')
-                    process_size = gr.Slider(minimum=64, maximum=2048, step=64, label="Size (width and height)", value=512)
+                    process_width = gr.Slider(minimum=64, maximum=2048, step=64, label="Width", value=512)
+                    process_height = gr.Slider(minimum=64, maximum=2048, step=64, label="Height", value=512)
 
                     with gr.Row():
                         process_flip = gr.Checkbox(label='Create flipped copies')
@@ -1050,7 +1051,8 @@ def create_ui(wrap_gradio_gpu_call):
                     dataset_directory = gr.Textbox(label='Dataset directory', placeholder="Path to directory with input images")
                     log_directory = gr.Textbox(label='Log directory', placeholder="Path to directory where to write outputs", value="textual_inversion")
                     template_file = gr.Textbox(label='Prompt template file', value=os.path.join(script_path, "textual_inversion_templates", "style_filewords.txt"))
-                    training_size = gr.Slider(minimum=64, maximum=2048, step=64, label="Size (width and height)", value=512)
+                    training_width = gr.Slider(minimum=64, maximum=2048, step=64, label="Width", value=512)
+                    training_height = gr.Slider(minimum=64, maximum=2048, step=64, label="Height", value=512)
                     steps = gr.Number(label='Max steps', value=100000, precision=0)
                     num_repeats = gr.Number(label='Number of repeats for a single input image per epoch', value=100, precision=0)
                     create_image_every = gr.Number(label='Save an image to log directory every N steps, 0 to disable', value=500, precision=0)
@@ -1095,7 +1097,8 @@ def create_ui(wrap_gradio_gpu_call):
             inputs=[
                 process_src,
                 process_dst,
-                process_size,
+                process_width,
+                process_height,
                 process_flip,
                 process_split,
                 process_caption,
@@ -1114,7 +1117,8 @@ def create_ui(wrap_gradio_gpu_call):
                 learn_rate,
                 dataset_directory,
                 log_directory,
-                training_size,
+                training_width,
+                training_height,
                 steps,
                 num_repeats,
                 create_image_every,
-- 
cgit v1.2.3


From bc3e183b739913e7be91213a256f038b10eb71e9 Mon Sep 17 00:00:00 2001
From: alg-wiki <alg.4chan@gmail.com>
Date: Tue, 11 Oct 2022 04:30:13 +0900
Subject: Textual Inversion: Preprocess and Training will only pick-up image
 files

---
 modules/textual_inversion/dataset.py           | 3 ++-
 modules/textual_inversion/preprocess.py        | 3 ++-
 modules/textual_inversion/textual_inversion.py | 3 ++-
 3 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py
index bcf772d2..d4baf066 100644
--- a/modules/textual_inversion/dataset.py
+++ b/modules/textual_inversion/dataset.py
@@ -22,6 +22,7 @@ class PersonalizedBase(Dataset):
         self.width = width
         self.height = height
         self.flip = transforms.RandomHorizontalFlip(p=flip_p)
+        self.extns = [".jpg",".jpeg",".png"]
 
         self.dataset = []
 
@@ -32,7 +33,7 @@ class PersonalizedBase(Dataset):
 
         assert data_root, 'dataset directory not specified'
 
-        self.image_paths = [os.path.join(data_root, file_path) for file_path in os.listdir(data_root)]
+        self.image_paths = [os.path.join(data_root, file_path) for file_path in os.listdir(data_root) if os.path.splitext(file_path.casefold())[1] in self.extns]
         print("Preparing dataset...")
         for path in tqdm.tqdm(self.image_paths):
             image = Image.open(path)
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index d7efdef2..b6c78cf8 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -12,12 +12,13 @@ def preprocess(process_src, process_dst, process_width, process_height, process_
     height = process_height
     src = os.path.abspath(process_src)
     dst = os.path.abspath(process_dst)
+    extns = [".jpg",".jpeg",".png"]
 
     assert src != dst, 'same directory specified as source and destination'
 
     os.makedirs(dst, exist_ok=True)
 
-    files = os.listdir(src)
+    files = [i for i in os.listdir(src) if os.path.splitext(i.casefold())[1] in extns]
 
     shared.state.textinfo = "Preprocessing..."
     shared.state.job_count = len(files)
diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py
index 5965c5a0..45397be9 100644
--- a/modules/textual_inversion/textual_inversion.py
+++ b/modules/textual_inversion/textual_inversion.py
@@ -161,6 +161,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
 
     shared.state.textinfo = "Initializing textual inversion training..."
     shared.state.job_count = steps
+    extns = [".jpg",".jpeg",".png"]
 
     filename = os.path.join(shared.cmd_opts.embeddings_dir, f'{embedding_name}.pt')
 
@@ -200,7 +201,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
     if ititial_step > steps:
         return embedding, filename
 
-    tr_img_len = len([os.path.join(data_root, file_path) for file_path in os.listdir(data_root)])
+    tr_img_len = len([os.path.join(data_root, file_path) for file_path in os.listdir(data_root) if os.path.splitext(file_path.casefold())[1] in extns])
     epoch_len = (tr_img_len * num_repeats) + tr_img_len
 
     pbar = tqdm.tqdm(enumerate(ds), total=steps-ititial_step)
-- 
cgit v1.2.3


From 907a88b2d0be320575c2129d8d6a1d4f3a68f9eb Mon Sep 17 00:00:00 2001
From: alg-wiki <alg.4chan@gmail.com>
Date: Tue, 11 Oct 2022 06:33:08 +0900
Subject: Added .webp .bmp

---
 modules/textual_inversion/dataset.py           | 2 +-
 modules/textual_inversion/preprocess.py        | 2 +-
 modules/textual_inversion/textual_inversion.py | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py
index d4baf066..0dc54fb7 100644
--- a/modules/textual_inversion/dataset.py
+++ b/modules/textual_inversion/dataset.py
@@ -22,7 +22,7 @@ class PersonalizedBase(Dataset):
         self.width = width
         self.height = height
         self.flip = transforms.RandomHorizontalFlip(p=flip_p)
-        self.extns = [".jpg",".jpeg",".png"]
+        self.extns = [".jpg",".jpeg",".png",".webp",".bmp"]
 
         self.dataset = []
 
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index b6c78cf8..8290abe8 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -12,7 +12,7 @@ def preprocess(process_src, process_dst, process_width, process_height, process_
     height = process_height
     src = os.path.abspath(process_src)
     dst = os.path.abspath(process_dst)
-    extns = [".jpg",".jpeg",".png"]
+    extns = [".jpg",".jpeg",".png",".webp",".bmp"]
 
     assert src != dst, 'same directory specified as source and destination'
 
diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py
index a03b299c..33c923d1 100644
--- a/modules/textual_inversion/textual_inversion.py
+++ b/modules/textual_inversion/textual_inversion.py
@@ -161,7 +161,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
 
     shared.state.textinfo = "Initializing textual inversion training..."
     shared.state.job_count = steps
-    extns = [".jpg",".jpeg",".png"]
+    extns = [".jpg",".jpeg",".png",".webp",".bmp"]
 
     filename = os.path.join(shared.cmd_opts.embeddings_dir, f'{embedding_name}.pt')
 
-- 
cgit v1.2.3


From bb932dbf9faf43ba918daa4791873078797b2a48 Mon Sep 17 00:00:00 2001
From: JC_Array <justin318@att.net>
Date: Mon, 10 Oct 2022 18:37:52 -0500
Subject: added alpha sort and threshold variables to create process method in
 preprocessing

---
 modules/textual_inversion/preprocess.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index 4a2194da..c0af729b 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -29,7 +29,7 @@ def preprocess(process_src, process_dst, process_width, process_height, process_
         shared.interrogator.load()
 
     if process_caption_deepbooru:
-        deepbooru.create_deepbooru_process()
+        deepbooru.create_deepbooru_process(opts.deepbooru_threshold, opts.deepbooru_sort_alpha)
 
     def save_pic_with_caption(image, index):
         if process_caption:
-- 
cgit v1.2.3


From b2368a3bce663f19a7209d9cb38617e635ca6e3c Mon Sep 17 00:00:00 2001
From: alg-wiki <alg.4chan@gmail.com>
Date: Tue, 11 Oct 2022 17:32:46 +0900
Subject: Switched to exception handling

---
 modules/textual_inversion/dataset.py           | 10 +++++-----
 modules/textual_inversion/preprocess.py        |  8 +++++---
 modules/textual_inversion/textual_inversion.py | 18 ++++++++----------
 3 files changed, 18 insertions(+), 18 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py
index 0dc54fb7..4d006366 100644
--- a/modules/textual_inversion/dataset.py
+++ b/modules/textual_inversion/dataset.py
@@ -22,7 +22,6 @@ class PersonalizedBase(Dataset):
         self.width = width
         self.height = height
         self.flip = transforms.RandomHorizontalFlip(p=flip_p)
-        self.extns = [".jpg",".jpeg",".png",".webp",".bmp"]
 
         self.dataset = []
 
@@ -33,12 +32,13 @@ class PersonalizedBase(Dataset):
 
         assert data_root, 'dataset directory not specified'
 
-        self.image_paths = [os.path.join(data_root, file_path) for file_path in os.listdir(data_root) if os.path.splitext(file_path.casefold())[1] in self.extns]
+        self.image_paths = [os.path.join(data_root, file_path) for file_path in os.listdir(data_root)]
         print("Preparing dataset...")
         for path in tqdm.tqdm(self.image_paths):
-            image = Image.open(path)
-            image = image.convert('RGB')
-            image = image.resize((self.width, self.height), PIL.Image.BICUBIC)
+            try:
+                image = Image.open(path).convert('RGB').resize((self.width, self.height), PIL.Image.BICUBIC)
+            except Exception:
+                continue
 
             filename = os.path.basename(path)
             filename_tokens = os.path.splitext(filename)[0]
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index 8290abe8..1a672725 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -12,13 +12,12 @@ def preprocess(process_src, process_dst, process_width, process_height, process_
     height = process_height
     src = os.path.abspath(process_src)
     dst = os.path.abspath(process_dst)
-    extns = [".jpg",".jpeg",".png",".webp",".bmp"]
 
     assert src != dst, 'same directory specified as source and destination'
 
     os.makedirs(dst, exist_ok=True)
 
-    files = [i for i in os.listdir(src) if os.path.splitext(i.casefold())[1] in extns]
+    files = os.listdir(src)
 
     shared.state.textinfo = "Preprocessing..."
     shared.state.job_count = len(files)
@@ -47,7 +46,10 @@ def preprocess(process_src, process_dst, process_width, process_height, process_
     for index, imagefile in enumerate(tqdm.tqdm(files)):
         subindex = [0]
         filename = os.path.join(src, imagefile)
-        img = Image.open(filename).convert("RGB")
+        try:
+            img = Image.open(filename).convert("RGB")
+        except Exception:
+            continue
 
         if shared.state.interrupted:
             break
diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py
index 33c923d1..91cde04b 100644
--- a/modules/textual_inversion/textual_inversion.py
+++ b/modules/textual_inversion/textual_inversion.py
@@ -161,7 +161,6 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
 
     shared.state.textinfo = "Initializing textual inversion training..."
     shared.state.job_count = steps
-    extns = [".jpg",".jpeg",".png",".webp",".bmp"]
 
     filename = os.path.join(shared.cmd_opts.embeddings_dir, f'{embedding_name}.pt')
 
@@ -201,10 +200,6 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
     if ititial_step > steps:
         return embedding, filename
 
-    tr_img_len = len([os.path.join(data_root, file_path) for file_path in os.listdir(data_root) if os.path.splitext(file_path.casefold())[1] in extns])
-
-    epoch_len = (tr_img_len * num_repeats) + tr_img_len
-
     pbar = tqdm.tqdm(enumerate(ds), total=steps-ititial_step)
     for i, (x, text) in pbar:
         embedding.step = i + ititial_step
@@ -228,10 +223,10 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
             loss.backward()
             optimizer.step()
 
-        epoch_num = embedding.step // epoch_len
-        epoch_step = embedding.step - (epoch_num * epoch_len) + 1
+        epoch_num = embedding.step // len(ds)
+        epoch_step = embedding.step - (epoch_num * len(ds)) + 1
 
-        pbar.set_description(f"[Epoch {epoch_num}: {epoch_step}/{epoch_len}]loss: {losses.mean():.7f}")
+        pbar.set_description(f"[Epoch {epoch_num}: {epoch_step}/{len(ds)}]loss: {losses.mean():.7f}")
 
         if embedding.step > 0 and embedding_dir is not None and embedding.step % save_embedding_every == 0:
             last_saved_file = os.path.join(embedding_dir, f'{embedding_name}-{embedding.step}.pt')
@@ -243,9 +238,12 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini
             p = processing.StableDiffusionProcessingTxt2Img(
                 sd_model=shared.sd_model,
                 prompt=text,
-                steps=20,
-				height=training_height,
+                steps=28,
+				height=768,
 				width=training_width,
+                negative_prompt="lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts,signature, watermark, username, blurry, artist name",
+                cfg_scale=7.0,
+                sampler_index=0,
                 do_not_save_grid=True,
                 do_not_save_samples=True,
             )
-- 
cgit v1.2.3


From f53f703aebc801c4204182d52bb1e0bef9808e1f Mon Sep 17 00:00:00 2001
From: JC_Array <justin318@att.net>
Date: Tue, 11 Oct 2022 18:12:12 -0500
Subject: resolved conflicts, moved settings under interrogate section,
 settings only show if deepbooru flag is enabled

---
 modules/deepbooru.py                    |  2 +-
 modules/shared.py                       | 19 +++++++++----------
 modules/textual_inversion/preprocess.py |  2 +-
 modules/ui.py                           |  2 +-
 4 files changed, 12 insertions(+), 13 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/deepbooru.py b/modules/deepbooru.py
index 89dcac3c..29529949 100644
--- a/modules/deepbooru.py
+++ b/modules/deepbooru.py
@@ -8,7 +8,7 @@ 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
-    create_deepbooru_process(shared.opts.deepbooru_threshold, shared.opts.deepbooru_sort_alpha)
+    create_deepbooru_process(shared.opts.interrogate_deepbooru_score_threshold, shared.opts.deepbooru_sort_alpha)
     shared.deepbooru_process_return["value"] = -1
     shared.deepbooru_process_queue.put(pil_image)
     while shared.deepbooru_process_return["value"] == -1:
diff --git a/modules/shared.py b/modules/shared.py
index 817203f8..5456c477 100644
--- a/modules/shared.py
+++ b/modules/shared.py
@@ -248,15 +248,20 @@ options_templates.update(options_section(('sd', "Stable Diffusion"), {
     "random_artist_categories": OptionInfo([], "Allowed categories for random artists selection when using the Roll button", gr.CheckboxGroup, {"choices": artist_db.categories()}),
 }))
 
-options_templates.update(options_section(('interrogate', "Interrogate Options"), {
+interrogate_option_dictionary = {
     "interrogate_keep_models_in_memory": OptionInfo(False, "Interrogate: keep models in VRAM"),
     "interrogate_use_builtin_artists": OptionInfo(True, "Interrogate: use artists from artists.csv"),
     "interrogate_clip_num_beams": OptionInfo(1, "Interrogate: num_beams for BLIP", gr.Slider, {"minimum": 1, "maximum": 16, "step": 1}),
     "interrogate_clip_min_length": OptionInfo(24, "Interrogate: minimum description length (excluding artists, etc..)", gr.Slider, {"minimum": 1, "maximum": 128, "step": 1}),
     "interrogate_clip_max_length": OptionInfo(48, "Interrogate: maximum description length", gr.Slider, {"minimum": 1, "maximum": 256, "step": 1}),
-    "interrogate_clip_dict_limit": OptionInfo(1500, "Interrogate: maximum number of lines in text file (0 = No limit)"),
-    "interrogate_deepbooru_score_threshold": OptionInfo(0.5, "Interrogate: deepbooru score threshold", gr.Slider, {"minimum": 0, "maximum": 1, "step": 0.01}),
-}))
+    "interrogate_clip_dict_limit": OptionInfo(1500, "Interrogate: maximum number of lines in text file (0 = No limit)")
+}
+
+if cmd_opts.deepdanbooru:
+    interrogate_option_dictionary["interrogate_deepbooru_score_threshold"] = OptionInfo(0.5, "Interrogate: deepbooru score threshold", gr.Slider, {"minimum": 0, "maximum": 1, "step": 0.01})
+    interrogate_option_dictionary["deepbooru_sort_alpha"] = OptionInfo(True, "Interrogate: deepbooru sort alphabetically", gr.Checkbox)
+
+options_templates.update(options_section(('interrogate', "Interrogate Options"), interrogate_option_dictionary))
 
 options_templates.update(options_section(('ui', "User interface"), {
     "show_progressbar": OptionInfo(True, "Show progressbar"),
@@ -282,12 +287,6 @@ options_templates.update(options_section(('sampler-params', "Sampler parameters"
     'eta_noise_seed_delta': OptionInfo(0, "Eta noise seed delta", gr.Number, {"precision": 0}),
 }))
 
-if cmd_opts.deepdanbooru:
-    options_templates.update(options_section(('deepbooru-params', "DeepBooru parameters"), {
-        "deepbooru_sort_alpha": OptionInfo(True, "Sort Alphabetical", gr.Checkbox),
-        'deepbooru_threshold': OptionInfo(0.5, "Threshold", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.01}),
-    }))
-
 
 class Options:
     data = None
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index a96388d6..113cecf1 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -29,7 +29,7 @@ def preprocess(process_src, process_dst, process_width, process_height, process_
         shared.interrogator.load()
 
     if process_caption_deepbooru:
-        deepbooru.create_deepbooru_process(opts.deepbooru_threshold, opts.deepbooru_sort_alpha)
+        deepbooru.create_deepbooru_process(opts.interrogate_deepbooru_score_threshold, opts.deepbooru_sort_alpha)
 
     def save_pic_with_caption(image, index):
         if process_caption:
diff --git a/modules/ui.py b/modules/ui.py
index 2891fc8c..fa45edca 100644
--- a/modules/ui.py
+++ b/modules/ui.py
@@ -317,7 +317,7 @@ def interrogate(image):
 
 
 def interrogate_deepbooru(image):
-    prompt = get_deepbooru_tags(image, opts.interrogate_deepbooru_score_threshold)
+    prompt = get_deepbooru_tags(image)
     return gr_show(True) if prompt is None else prompt
 
 
-- 
cgit v1.2.3


From 698d303b04e293635bfb49c525409f3bcf671dce Mon Sep 17 00:00:00 2001
From: AUTOMATIC <16777216c@gmail.com>
Date: Wed, 12 Oct 2022 21:55:43 +0300
Subject: deepbooru: added option to use spaces or underscores deepbooru: added
 option to quote (\) in tags deepbooru/BLIP: write caption to file instead of
 image filename deepbooru/BLIP: now possible to use both for captions
 deepbooru: process is stopped even if an exception occurs

---
 modules/deepbooru.py                    | 65 ++++++++++++++++++-----
 modules/shared.py                       |  2 +
 modules/textual_inversion/preprocess.py | 92 ++++++++++++++-------------------
 modules/ui.py                           |  7 +--
 4 files changed, 95 insertions(+), 71 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/deepbooru.py b/modules/deepbooru.py
index 29529949..419e6a9c 100644
--- a/modules/deepbooru.py
+++ b/modules/deepbooru.py
@@ -2,33 +2,44 @@ import os.path
 from concurrent.futures import ProcessPoolExecutor
 import multiprocessing
 import time
+import re
+
+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
-    create_deepbooru_process(shared.opts.interrogate_deepbooru_score_threshold, shared.opts.deepbooru_sort_alpha)
-    shared.deepbooru_process_return["value"] = -1
-    shared.deepbooru_process_queue.put(pil_image)
-    while shared.deepbooru_process_return["value"] == -1:
-        time.sleep(0.2)
-    tags = shared.deepbooru_process_return["value"]
-    release_process()
-    return tags
 
+    try:
+        create_deepbooru_process(shared.opts.interrogate_deepbooru_score_threshold, create_deepbooru_opts())
+        return get_tags_from_process(pil_image)
+    finally:
+        release_process()
+
+
+def create_deepbooru_opts():
+    from modules import shared
 
-def deepbooru_process(queue, deepbooru_process_return, threshold, alpha_sort):
+    return {
+        "use_spaces": shared.opts.deepbooru_use_spaces,
+        "use_escape": shared.opts.deepbooru_escape,
+        "alpha_sort": shared.opts.deepbooru_sort_alpha,
+    }
+
+
+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, alpha_sort)
+            deepbooru_process_return["value"] = get_deepbooru_tags_from_model(model, tags, pil_image, threshold, deepbooru_opts)
 
 
-def create_deepbooru_process(threshold, alpha_sort):
+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
@@ -41,10 +52,23 @@ def create_deepbooru_process(threshold, alpha_sort):
     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 = multiprocessing.Process(target=deepbooru_process, args=(shared.deepbooru_process_queue, shared.deepbooru_process_return, threshold, alpha_sort))
+    shared.deepbooru_process = multiprocessing.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
@@ -81,10 +105,15 @@ def get_deepbooru_tags_model():
     return model, tags
 
 
-def get_deepbooru_tags_from_model(model, tags, pil_image, threshold, alpha_sort):
+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']
+
     width = model.input_shape[2]
     height = model.input_shape[1]
     image = np.array(pil_image)
@@ -129,4 +158,12 @@ def get_deepbooru_tags_from_model(model, tags, pil_image, threshold, alpha_sort)
 
     print('\n'.join(sorted(result_tags_print, reverse=True)))
 
-    return ', '.join(result_tags_out).replace('_', ' ').replace(':', ' ')
+    tags_text = ', '.join(result_tags_out)
+
+    if use_spaces:
+        tags_text = tags_text.replace('_', ' ')
+
+    if use_escape:
+        tags_text = re.sub(re_special, r'\\\1', tags_text)
+
+    return tags_text.replace(':', ' ')
diff --git a/modules/shared.py b/modules/shared.py
index e64e69fc..78b73aae 100644
--- a/modules/shared.py
+++ b/modules/shared.py
@@ -260,6 +260,8 @@ options_templates.update(options_section(('interrogate', "Interrogate Options"),
     "interrogate_clip_max_length": OptionInfo(48, "Interrogate: maximum description length", gr.Slider, {"minimum": 1, "maximum": 256, "step": 1}),
     "interrogate_deepbooru_score_threshold": OptionInfo(0.5, "Interrogate: deepbooru score threshold", gr.Slider, {"minimum": 0, "maximum": 1, "step": 0.01}),
     "deepbooru_sort_alpha": OptionInfo(True, "Interrogate: deepbooru sort alphabetically"),
+    "deepbooru_use_spaces": OptionInfo(False, "use spaces for tags in deepbooru"),
+    "deepbooru_escape": OptionInfo(True, "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)"),
 }))
 
 options_templates.update(options_section(('ui', "User interface"), {
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index 113cecf1..3047bede 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -10,7 +10,28 @@ from modules.shared import opts, cmd_opts
 if cmd_opts.deepdanbooru:
     import modules.deepbooru as deepbooru
 
+
 def preprocess(process_src, process_dst, process_width, process_height, process_flip, process_split, process_caption, process_caption_deepbooru=False):
+    try:
+        if process_caption:
+            shared.interrogator.load()
+
+        if process_caption_deepbooru:
+            deepbooru.create_deepbooru_process(opts.interrogate_deepbooru_score_threshold, deepbooru.create_deepbooru_opts())
+
+        preprocess_work(process_src, process_dst, process_width, process_height, process_flip, process_split, process_caption, process_caption_deepbooru)
+
+    finally:
+
+        if process_caption:
+            shared.interrogator.send_blip_to_ram()
+
+        if process_caption_deepbooru:
+            deepbooru.release_process()
+
+
+
+def preprocess_work(process_src, process_dst, process_width, process_height, process_flip, process_split, process_caption, process_caption_deepbooru=False):
     width = process_width
     height = process_height
     src = os.path.abspath(process_src)
@@ -25,30 +46,28 @@ def preprocess(process_src, process_dst, process_width, process_height, process_
     shared.state.textinfo = "Preprocessing..."
     shared.state.job_count = len(files)
 
-    if process_caption:
-        shared.interrogator.load()
-
-    if process_caption_deepbooru:
-        deepbooru.create_deepbooru_process(opts.interrogate_deepbooru_score_threshold, opts.deepbooru_sort_alpha)
-
     def save_pic_with_caption(image, index):
+        caption = ""
+
         if process_caption:
-            caption = "-" + shared.interrogator.generate_caption(image)
-            caption = sanitize_caption(os.path.join(dst, f"{index:05}-{subindex[0]}"), caption, ".png")
-        elif process_caption_deepbooru:
-            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"]
-            caption = sanitize_caption(os.path.join(dst, f"{index:05}-{subindex[0]}"), caption, ".png")
-            shared.deepbooru_process_return["value"] = -1
-        else:
-            caption = filename
-            caption = os.path.splitext(caption)[0]
-            caption = os.path.basename(caption)
+            caption += shared.interrogator.generate_caption(image)
+
+        if process_caption_deepbooru:
+            if len(caption) > 0:
+                caption += ", "
+            caption += deepbooru.get_tags_from_process(image)
+
+        filename_part = filename
+        filename_part = os.path.splitext(filename_part)[0]
+        filename_part = os.path.basename(filename_part)
+
+        basename = f"{index:05}-{subindex[0]}-{filename_part}"
+        image.save(os.path.join(dst, f"{basename}.png"))
+
+        if len(caption) > 0:
+            with open(os.path.join(dst, f"{basename}.txt"), "w", encoding="utf8") as file:
+                file.write(caption)
 
-        image.save(os.path.join(dst, f"{index:05}-{subindex[0]}{caption}.png"))
         subindex[0] += 1
 
     def save_pic(image, index):
@@ -93,34 +112,3 @@ def preprocess(process_src, process_dst, process_width, process_height, process_
             save_pic(img, index)
 
         shared.state.nextjob()
-
-    if process_caption:
-        shared.interrogator.send_blip_to_ram()
-
-    if process_caption_deepbooru:
-        deepbooru.release_process()
-
-
-def sanitize_caption(base_path, original_caption, suffix):
-    operating_system = platform.system().lower()
-    if (operating_system == "windows"):
-        invalid_path_characters = "\\/:*?\"<>|"
-        max_path_length = 259
-    else:
-        invalid_path_characters = "/" #linux/macos
-        max_path_length = 1023
-    caption = original_caption
-    for invalid_character in invalid_path_characters:
-        caption = caption.replace(invalid_character, "")
-    fixed_path_length = len(base_path) + len(suffix) 
-    if fixed_path_length + len(caption) <= max_path_length:
-        return caption
-    caption_tokens = caption.split()
-    new_caption = ""
-    for token in caption_tokens:
-        last_caption = new_caption
-        new_caption = new_caption + token + " "
-        if (len(new_caption) + fixed_path_length - 1  > max_path_length):
-            break
-    print(f"\nPath will be too long. Truncated caption: {original_caption}\nto: {last_caption}", file=sys.stderr)
-    return last_caption.strip()
diff --git a/modules/ui.py b/modules/ui.py
index c42535c8..e07ee0e1 100644
--- a/modules/ui.py
+++ b/modules/ui.py
@@ -1074,11 +1074,8 @@ def create_ui(wrap_gradio_gpu_call):
                     with gr.Row():
                         process_flip = gr.Checkbox(label='Create flipped copies')
                         process_split = gr.Checkbox(label='Split oversized images into two')
-                        process_caption = gr.Checkbox(label='Use BLIP caption as filename')
-                        if cmd_opts.deepdanbooru:
-                            process_caption_deepbooru = gr.Checkbox(label='Use deepbooru caption as filename')
-                        else:
-                            process_caption_deepbooru = gr.Checkbox(label='Use deepbooru caption as filename', visible=False)
+                        process_caption = gr.Checkbox(label='Use BLIP for caption')
+                        process_caption_deepbooru = gr.Checkbox(label='Use deepbooru for caption', visible=True if cmd_opts.deepdanbooru else False)
 
                     with gr.Row():
                         with gr.Column(scale=3):
-- 
cgit v1.2.3


From f776254b12361b5bae16f6629bcdcb47b450c48d Mon Sep 17 00:00:00 2001
From: Greg Fuller <gfuller23@gmail.com>
Date: Wed, 12 Oct 2022 13:08:06 -0700
Subject: [2/?] [wip] ignore OPT_INCLUDE_RANKS for training filenames

---
 modules/deepbooru.py                    | 3 ++-
 modules/textual_inversion/preprocess.py | 4 +++-
 2 files changed, 5 insertions(+), 2 deletions(-)

(limited to 'modules/textual_inversion/preprocess.py')

diff --git a/modules/deepbooru.py b/modules/deepbooru.py
index 2cbf2cab..fcc05819 100644
--- a/modules/deepbooru.py
+++ b/modules/deepbooru.py
@@ -19,6 +19,7 @@ def get_deepbooru_tags(pil_image):
         release_process()
 
 
+OPT_INCLUDE_RANKS = "include_ranks"
 def create_deepbooru_opts():
     from modules import shared
 
@@ -26,7 +27,7 @@ def create_deepbooru_opts():
         "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,
+        OPT_INCLUDE_RANKS: shared.opts.interrogate_return_ranks,
     }
 
 
diff --git a/modules/textual_inversion/preprocess.py b/modules/textual_inversion/preprocess.py
index 3047bede..886cf0c3 100644
--- a/modules/textual_inversion/preprocess.py
+++ b/modules/textual_inversion/preprocess.py
@@ -17,7 +17,9 @@ def preprocess(process_src, process_dst, process_width, process_height, process_
             shared.interrogator.load()
 
         if process_caption_deepbooru:
-            deepbooru.create_deepbooru_process(opts.interrogate_deepbooru_score_threshold, deepbooru.create_deepbooru_opts())
+            db_opts = deepbooru.create_deepbooru_opts()
+            db_opts[deepbooru.OPT_INCLUDE_RANKS] = False
+            deepbooru.create_deepbooru_process(opts.interrogate_deepbooru_score_threshold, db_opts)
 
         preprocess_work(process_src, process_dst, process_width, process_height, process_flip, process_split, process_caption, process_caption_deepbooru)
 
-- 
cgit v1.2.3