LarryAIDraw commited on
Commit
6c8ba7a
·
1 Parent(s): e324c8b

Upload 5 files

Browse files
ashigara_kantaicollection1040.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e016b44e8eaf1d52debef67fed03980e74e2737731fb6e60e23ed6c6722656c
3
+ size 13296
ashigara_kantaicollection1040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:064b1bb90d66d0e0899e30da9896e323d45a872f6eeab1fb93cbdf0756f9342b
3
+ size 14722160
ashigara_kantaicollection520.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a93b16cc441647202f5d105f2a80699bf083de04cf296c2d4b0abd9255241b1
3
+ size 13293
ashigara_kantaicollection520.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7962ddf7ecd0993daa9f3dd3747a14ca86fba43e8dbe264f136ed08bf2538df8
3
+ size 14722160
ashigara_足柄 (Kantai Collection) - v1.4 Stable Diffusion LoRA Civitai.txt ADDED
@@ -0,0 +1,438 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ santa hat
2
+ deerdeer nosedeer glow
3
+ <https://civitai.com/models>
4
+
5
+ <https://civitai.com/pricing>
6
+
7
+ /
8
+
9
+ Create
10
+
11
+ larry110864584's Avatar
12
+
13
+ 0
14
+
15
+ Create
16
+
17
+
18
+ ashigara/足柄 (Kantai Collection)
19
+
20
+ 59
21
+ 188
22
+ 5
23
+ 0
24
+
25
+ 1
26
+
27
+ Updated: Nov 19, 2023
28
+ character <https://civitai.com/tag/character>
29
+ female <https://civitai.com/tag/female>brown eyes
30
+ <https://civitai.com/tag/brown%20eyes>breasts
31
+ <https://civitai.com/tag/breasts>long hair
32
+ <https://civitai.com/tag/long%20hair>smile <https://civitai.com/tag/smile>
33
+ + 12
34
+
35
+ v1.4
36
+
37
+ pattern_2.png
38
+ <https://civitai.com/images/3750062>
39
+
40
+ 👍
41
+ 2
42
+
43
+ ❤️
44
+ 3
45
+
46
+ 0
47
+
48
+ pattern_6.png
49
+ <https://civitai.com/images/3750063>
50
+
51
+ 👍
52
+ 1
53
+
54
+ ❤️
55
+ 1
56
+
57
+ 0
58
+
59
+ suit.png
60
+ <https://civitai.com/images/3750064>
61
+
62
+ 0
63
+
64
+ yukata.png
65
+ <https://civitai.com/images/3750065>
66
+
67
+ 0
68
+
69
+ 13
70
+
71
+ <https://civitai.com/images/3750068>
72
+ 👍
73
+ 1
74
+
75
+ ❤️
76
+ 1
77
+
78
+ 0
79
+
80
+ maid.png
81
+ <https://civitai.com/images/3750069>
82
+
83
+ 0
84
+
85
+ miko.png
86
+ <https://civitai.com/images/3750071>
87
+
88
+ 0
89
+
90
+ Create
91
+
92
+ Verified: 10 days ago
93
+ SafeTensor
94
+
95
+ Details
96
+
97
+ Type
98
+
99
+ LoRA
100
+ <https://github.com/civitai/civitai/wiki/How-to-use-models#lora>
101
+ Downloads
102
+ 188
103
+ Uploaded
104
+ Nov 19, 2023
105
+ Base Model
106
+
107
+ SD 1.5
108
+ Training
109
+
110
+ Steps: 4,680
111
+ Epochs: 10
112
+ Usage Tips
113
+
114
+ Clip Skip: 2
115
+ Trigger Words
116
+
117
+ ashigara_kantaicollection
118
+
119
+ long_hair, brown_hair, brown_eyes, hairband, breasts, smile, open_mouth,
120
+ wavy_hair, blush, large_breasts, hair_between_eyes
121
+
122
+ Hash
123
+
124
+ AutoV2
125
+ F799CDA713
126
+
127
+ 2 Files
128
+
129
+ LoRA (12.98 KB)
130
+ Download
131
+ <https://civitai.com/api/download/models/231530?type=Model&format=PickleTensor>
132
+
133
+ Verified: 10 days ago
134
+ PickleTensor
135
+
136
+ LoRA (14.04 MB)
137
+ Download
138
+ <https://civitai.com/api/download/models/231530?type=Model&format=SafeTensor>
139
+
140
+ Verified: 10 days ago
141
+ SafeTensor
142
+
143
+ Reviews
144
+ 1 version ratings
145
+
146
+ 5 out of 5
147
+ Add Review
148
+ <https://civitai.com/posts/create?modelId=205641&modelVersionId=231530&reviewing=true&returnUrl=/models/205641/ashigara-kantai-collection>See Reviews <https://civitai.com/models/205641/reviews?modelVersionId=231530>
149
+
150
+ 5 star
151
+ 100%
152
+ 4 star
153
+ 0%
154
+ 3 star
155
+ 0%
156
+ 2 star
157
+ 0%
158
+ 1 star
159
+ 0%
160
+
161
+ About this version
162
+
163
+ narugo1992's Avatar
164
+ <https://civitai.com/user/narugo1992>
165
+ narugo1992
166
+ <https://civitai.com/user/narugo1992>
167
+ Joined Jul 24, 2023
168
+
169
+ Tip
170
+
171
+ Follow
172
+
173
+ #19
174
+ <https://civitai.com/leaderboard/character?position=19>
175
+
176
+ 901
177
+ 725
178
+ 1.4k
179
+ 55k
180
+ 327k
181
+
182
+ <https://huggingface.co/CyberHarem>
183
+
184
+ <https://huggingface.co/BangumiBase>
185
+
186
+ <https://huggingface.co/deepghs>
187
+
188
+ <https://github.com/deepghs>
189
+
190
+ <https://github.com/narugo1992>
191
+
192
+ <https://tensor.art/u/644163821006558816>
193
+
194
+ pattern_2.png
195
+ <https://civitai.com/images/3750062>
196
+
197
+ 👍
198
+ 2
199
+
200
+ ❤️
201
+ 3
202
+
203
+ 0
204
+
205
+ pattern_6.png
206
+ <https://civitai.com/images/3750063>
207
+
208
+ 👍
209
+ 1
210
+
211
+ ❤️
212
+ 1
213
+
214
+ 0
215
+
216
+ suit.png
217
+ <https://civitai.com/images/3750064>
218
+
219
+ 0
220
+
221
+ yukata.png
222
+ <https://civitai.com/images/3750065>
223
+
224
+ 0
225
+
226
+ 13
227
+
228
+ <https://civitai.com/images/3750068>
229
+ 👍
230
+ 1
231
+
232
+ ❤️
233
+ 1
234
+
235
+ 0
236
+
237
+ maid.png
238
+ <https://civitai.com/images/3750069>
239
+
240
+ 0
241
+
242
+ miko.png
243
+ <https://civitai.com/images/3750071>
244
+
245
+ 0
246
+
247
+ * Thanks to Civitai's TOS, some images cannot be uploaded. *THE FULL
248
+ PREVIEW IMAGES CAN BE FOUND ON HUGGINGFACE
249
+ <https://huggingface.co/CyberHarem/ashigara_kantaicollection>*.
250
+ * *THIS MODEL HAS TWO FILES. YOU NEED TO USE THEM TOGETHER!!!*
251
+ * *The associated trigger words are only for reference, it may need to
252
+ be adjusted at some times*.
253
+ * Recommended weight of pt file is 0.5-1.0, weight of LoRA is 0.5-0.85.
254
+ * Images were generated using a few fixed prompts and dataset-based
255
+ clustered prompts. Random seeds were used, ruling out
256
+ cherry-picking. *What you see here is what you can get.*
257
+ * No specialized training was done for outfits. You can check our
258
+ provided preview post to get the prompts corresponding to the outfits.
259
+ * This model is trained with *513 images*.
260
+
261
+
262
+ How to Use This Model
263
+
264
+ *THIS MODEL HAS TWO FILES. YOU NEED TO USE THEM TOGETHER!!!*. In this
265
+ case, you need to download both |ashigara_kantaicollection.pt| and |ashigara_kantaicollection.safetensors|, then *use |ashigara_kantaicollection.pt| as texture inversion embedding, and use |ashigara_kantaicollection.safetensors| as LoRA at the same time*.
266
+
267
+ *このモデルには2つのファイルがあります。一緒に使う必要がありま
268
+ す!!!*。 この場合、|ashigara_kantaicollection.pt|と|ashigara_kantaicollection.safetensors|の両方をダウンロード する必要があります。|ashigara_kantaicollection.pt|をテクスチャ反転埋め込みとして使用し、同時に|ashigara_kantaicollection.safetensors|をLoRAとして使用してください。
269
+
270
+ *这个模型有两个文件。你需要同时使用它们!!!*。 在这种情况下,您需要下载|ashigara_kantaicollection.pt|和|ashigara_kantaicollection.safetensors|这两个文件,然后将|ashigara_kantaicollection.pt|用作纹理反转嵌入, 同时使用|ashigara_kantaicollection.safetensors|作为LoRA。
271
+
272
+ *이 모델은 두 개의 파일이 있습니다. 두 파일을 함께 사���해야 합니다!!!*.
273
+ 이 경우에는 |ashigara_kantaicollection.pt|와 |ashigara_kantaicollection.safetensors| 두 파일을 모두 다운로드하신 다음에 *|ashigara_kantaicollection.pt|을 텍스처 반전 임베딩으로 사용하고, 동시에 |ashigara_kantaicollection.safetensors|을 LoRA로 사용하셔야 합니다*.
274
+
275
+ (Translated with ChatGPT)
276
+
277
+ The trigger word is |ashigara_kantaicollection|, and the recommended tags are |best quality, masterpiece, highres, solo, {ashigara_kantaicollection:1.15}, long_hair, brown_hair, brown_eyes, hairband, breasts, smile, open_mouth, wavy_hair, blush, large_breasts, hair_between_eyes|.
278
+
279
+
280
+ How This Model Is Trained
281
+
282
+ This model is trained with HCP-Diffusion
283
+ <https://github.com/7eu7d7/HCP-Diffusion>. And the auto-training
284
+ framework is maintained by DeepGHS Team <https://huggingface.co/deepghs>.
285
+
286
+
287
+ Why Some Preview Images Not Look Like Ashigara Kantaicollection
288
+
289
+ *All the prompt texts* used on the preview images (which can be viewed
290
+ by clicking on the images) *are automatically generated using clustering
291
+ algorithms* based on feature information extracted from the training
292
+ dataset. The seed used during image generation is also randomly
293
+ generated, and the images have not undergone any selection or
294
+ modification. As a result, there is a possibility of the mentioned
295
+ issues occurring.
296
+
297
+ In practice, based on our internal testing, most models that experience
298
+ such issues perform better in actual usage than what is seen in the
299
+ preview images. *The only thing you may need to do is adjusting the tags
300
+ you are using*.
301
+
302
+
303
+ I Felt This Model May Be Overfitting or Underfitting, What Shall I Do
304
+
305
+ Our model has been published on huggingface repository -
306
+ CyberHarem/ashigara/kantaicollection/
307
+ <https://huggingface.co/CyberHarem/ashigara_kantaicollection>, where
308
+ models of all the steps are saved. Also, we published the training
309
+ dataset on huggingface dataset - CyberHarem/ashigarakantaicollection
310
+ <https://huggingface.co/datasets/CyberHarem/ashigara_kantaicollection>,
311
+ which may be helpful to you.
312
+
313
+
314
+ Why Not Just Using The Better-Selected Images
315
+
316
+ Our model's entire process, from data crawling, training, to generating
317
+ preview images and publishing, is *100% automated without any human
318
+ intervention*. It's an interesting experiment conducted by our team, and
319
+ for this purpose, we have developed a complete set of software
320
+ infrastructure, including data filtering, automatic training, and
321
+ automated publishing. Therefore, if possible, we would appreciate more
322
+ feedback or suggestions as they are highly valuable to us.
323
+
324
+
325
+ Why Can't the Desired Character Outfits Be Accurately Generated
326
+
327
+ Our current training data is sourced from various image websites, and
328
+ for a fully automated pipeline, it's challenging to accurately predict
329
+ which official images a character possesses. Consequently, outfit
330
+ generation relies on clustering based on labels from the training
331
+ dataset in an attempt to achieve the best possible recreation. We will
332
+ continue to address this issue and attempt optimization, but it remains
333
+ a challenge that cannot be completely resolved. The accuracy of outfit
334
+ recreation is also unlikely to match the level achieved by manually
335
+ trained models.
336
+
337
+ In fact, this model's greatest strengths lie in recreating the inherent
338
+ characteristics of the characters themselves and its relatively strong
339
+ generalization capabilities, owing to its larger dataset. As such, *this
340
+ model is well-suited for tasks such as changing outfits, posing
341
+ characters, and, of course, generating NSFW images of characters!*😉".
342
+
343
+ For the following groups, it is not recommended to use this model and we
344
+ express regret:
345
+
346
+ 1. Individuals who cannot tolerate any deviations from the original
347
+ character design, even in the slightest detail.
348
+ 2. Individuals who are facing the application scenarios with high
349
+ demands for accuracy in recreating character outfits.
350
+ 3. Individuals who cannot accept the potential randomness in
351
+ AI-generated images based on the Stable Diffusion algorithm.
352
+ 4. Individuals who are not comfortable with the fully automated process
353
+ of training character models using LoRA, or those who believe that
354
+ training character models must be done purely through manual
355
+ operations to avoid disrespecting the characters.
356
+ 5. Individuals who finds the generated image content offensive to their
357
+ values.
358
+
359
+ Show More
360
+
361
+
362
+ Discussion
363
+
364
+ Add Comment
365
+
366
+ There are no comments for this model yet.
367
+ Be the first to let the people know about this model by leaving your
368
+ comment.
369
+
370
+
371
+ Gallery
372
+
373
+ Add Post
374
+
375
+ Add Review
376
+
377
+ Newest
378
+
379
+ All Time
380
+
381
+ TK31's Avatar
382
+ <https://civitai.com/user/TK31>
383
+ TK31
384
+ <https://civitai.com/user/TK31>
385
+ 13 hours ago - v1.4
386
+
387
+ 5.0
388
+ <https://civitai.com/models/205641/ashigara-kantai-collection?dialog=resourceReview&reviewId=417289>
389
+
390
+ D816AC52AFAEDD1D9F4232DB510E88A94C28994D7CB05B6DF6A309565F03FD47.jpg
391
+ <https://civitai.com/images/4012147>
392
+
393
+ 0
394
+
395
+ narugo1992's Avatar
396
+ <https://civitai.com/user/narugo1992>
397
+ narugo1992
398
+ <https://civitai.com/user/narugo1992>
399
+ 10 days ago - v1.4
400
+
401
+ miko.png <https://civitai.com/images/3750071>
402
+
403
+ 0
404
+
405
+ © Civitai 2023
406
+ Support Us ❤️
407
+ <https://civitai.com/pricing>
408
+ Join Us 💼
409
+ <https://civitai.com/content/careers>
410
+ Terms of Service
411
+ <https://civitai.com/content/tos>
412
+ Privacy
413
+ <https://civitai.com/content/privacy>
414
+ API
415
+ <https://civitai.com/github/wiki/REST-API-Reference>
416
+ Status
417
+ <https://status.civitai.com/>
418
+ Newsroom
419
+ <https://civitai.com/newsroom>
420
+
421
+ <https://civitai.com/discord>
422
+
423
+ <https://civitai.com/twitter>
424
+
425
+ <https://civitai.com/instagram>
426
+
427
+ <https://civitai.com/youtube>
428
+
429
+ <https://civitai.com/tiktok>
430
+
431
+ <https://civitai.com/reddit>
432
+
433
+ <https://civitai.com/github>
434
+ 🪲 Bugs
435
+ <https://civitai.com/bugs>
436
+ 💡 Feature Requests
437
+ <https://civitai.com/feedback>
438
+