<h1 id="important">Important</h1>The lack of prompt information here makes this release a bit premature. But I figured since I was going to be testing this for a while I might as well put it up.Also, the Flux version is...rough. The Flux build is best used as an additional danganronpa-flavored guide to other loras rather than something to use on its own. Or at least that's my initial take. That said, I think I judged it a bit too quickly and too harshly. If you use it just make sure to read the Flux notes section for tips. I'd generally recomend the chroma lora over the flux version.<h1 id="what-is-this-in-theory">What is this in theory?</h1>An attempt to train a lora on every ultimate (and most significant non or post graduation ultimates) from the Danganronpa franchise with some students from The Hundred Line: Last Defense Academy as well as Shinigami and Pucci from Rain Code added in as well. Basically a big pile of Kodaka and Komatsuzaki's combined works. I also have a few characters from the Danganronpa Another fan games.I also wanted to try boosting how well single-shot gens of of multiple characters would do. So I specifically trained on some images of two or three characters with thematic or canon links. Haven't had a chance to test how well that worked yet. I'd assume that the more represented in the larger danbooru data the characters are together the better this would work out. So Hajime and Chiaki are probably going to go well. Tsumugi and Gonta not as much.<h1 id="what-is-this-in-practice">What is this in practice?</h1>No idea. I really just decided to try it out because I was curious what would happen and I had 'just' enough spare buzz to do it. So far testing has it working better than I'd feared but not as well as I'd hoped. But given the scope it's way too early to say. If I try again I think just doubling the dataset size would be enough to make this viable. Which I'll probably try once I've had time to collect and caption all the 100 line character sprites from the game.Issues found so far.There should be a couple images of Darumui from 100 line in the main gallery to highlight the problem. It 'mostly' gets her in the hemoanima form. But it still misses on some details that differentiate her from any other anime girl with gradient hair wearing a dark uniform. And even more so with the standard look. I'm assuming that's going to be how it'll go for a lot or all of the 100 line characters. The Danganronpa cast 'usually' had a strong presence in the Illustrious training data anyway. But here things pretty much begin and end with this lora training on a very small amount of imaes per character. It's the difference between just strengthening what's already there and adding something new. Basically the 100 line characters I trained on are in there and they 'kinda' work but you're better off just looking for a dedicated lora for them.<h1 id="usage-tips">Usage tips</h1>The ideal would be to mix it with a specific character lora. But if you're going kind of scattershot with prompts, an alternative would be to go for loras that 'also' have a wide scope of danganronpa-related training. My Tsumugi beta loras have a fair amount of non-tsumugi stuff in there too. And I tried to get a very wide range of characters in my danganronpa anime lora. So right now I've been playing around with a mix of this lora in upper mid range strength, the danganronpa anime lora at full strength, and the Tsumugi experimental lora (v2) at very low strength. This is a bit of a mess to be sure, but here's my current base prompt with all the loras:<h3 id="illustrious:">Illustrious:</h3>&lt;lora:Danganronpa_Locations:0.1&gt;, &lt;lora:Tsumugi_Shirogane_Danganronpa_Illustrious_V2:0.2&gt;, &lt;lora:Danganronpa_Anime_Style_Mix_V2_Illustrious:1&gt;, &lt;lora:100danganpack-illustrious:0.6&gt;, 100danganpack, danganronpa_3_(anime), danganronpa_3_(anime)_(style), danganronpa_(series),anime_screenshot , anime screencap, anime coloring, official_art, komatsuzaki_rui_(style) , YOUR_ACTUAL_PROMPT_HEREOr to push it even further with a bit more booster from ccdanganart:&lt;lora:Danganronpa_Locations:0.1&gt;, &lt;lora:Tsumugi_Shirogane_Danganronpa_Illustrious_V2:0.2&gt;, &lt;lora:ccdanganart:0.1&gt;, ccdanganart, &lt;lora:Danganronpa_Anime_Style_Mix_V2_Illustrious:1&gt;, &lt;lora:100danganpack-illustrious:0.6&gt;, 100danganpack, danganronpa_3_(anime), danganronpa_3_(anime)_(style), danganronpa_(series),anime_screenshot , anime screencap, anime coloring, official_art, komatsuzaki_rui_(style) , YOUR_ACTUAL_PROMPT_HERE<h3 id="flux-notes:">Flux Notes:</h3>The Flux version is finicky to the point where it's something of a failure, but one that's still useful as a crutch for other loras. I'm liking the style I'm getting with a mix of this lora, hyperflux, the danganronpa anime style, the nononononoway style, and finally a specific character lora with these strengths - using Tsumugi as an example. This pushes to a somewhat anime'ified version of the art style from the games. It's not the sprite style, not the anime style, but an interesting mix.&lt;lora:Hyper-FLUX.1-dev-8steps-lora:0.125&gt;, &lt;lora:100danganpack-flux:0.5&gt; , 100danganpack, &lt;lora:Danganronpa_Anime_Style_Mix_Flux_V2.01a:0.8&gt;, &lt;lora:Nononononoway_Style_Flux:0.9&gt;, &lt;lora:Tsumugi_Shirogane_Danganronpa_flux_tags_V2.01_25e:0.8&gt;, shirogane_tsumugi, A detailed anime-style professional illustration which features high quality anime digital art combined with detailed character design. Tsumugi Shirogane is wearing her uniform from Danganronpa V3. Tsumugi Shirogane has knee-length teal-blue hair parted in the center which is straight and smooth at the base and gradually splits into waves near the tips. Her eyes are large, round, and are a soft teal-green color. Tsumugi wears round, rimless glasses with white temples. Tsumugi's Danganronpa V3 school uniform consists of a white blouse with an orange bow tied at her collar underneath a two-button, pocketed, teal-black school blazer adorned with two silver buttons on each lapel and a single button on each gorge. She wears a knee-length teal-black pleated skirt with two dark teal stripes lining the hem. Her legwear consists of dark navy knee-length socks and teal-gray loafers. YOUR ACTUAL PROMPTRemoving the danganronpa anime lora, ironically, makes it 'more' anime by removing some of Rui's influence and letting the others compensate. This is one of my favorite variations where the Nononononoway lora plays a stronger role:&lt;lora:Hyper-FLUX.1-dev-8steps-lora:0.125&gt;, &lt;lora:100danganpack-flux:0.5&gt; , 100danganpack, &lt;lora:Nononononoway_Style_Flux:0.9&gt;, &lt;lora:Tsumugi_Shirogane_Danganronpa_flux_tags_V2.01_25e:0.8&gt;, A detailed anime-style professional illustration which features high quality anime digital art combined with detailed character design. Tsumugi Shirogane is wearing her uniform from Danganronpa V3. Tsumugi Shirogane has knee-length teal-blue hair parted in the center which is straight and smooth at the base and gradually splits into waves near the tips. Her eyes are large, round, and are a soft teal-green color. Tsumugi wears round, rimless glasses with white temples. Tsumugi's Danganronpa V3 school uniform consists of a white blouse with an orange bow tied at her collar underneath a two-button, pocketed, teal-black school blazer adorned with two silver buttons on each lapel and a single button on each gorge. She wears a knee-length teal-black pleated skirt with two dark teal stripes lining the hem. Her legwear consists of dark navy knee-length socks and teal-gray loafers. YOUR ACTUAL PROMPTThe downside of making a lora stew of course is that with every extra lora things are all getting further dilluted into each other. Which is a pain if you want a specific art style like the danganronpa anime. Though some with cross-lora training, like using komatsuzaki_rui_(style) , won't suffer from that nearly as much. However, even then there's different examples of komatsuzaki_rui_(style) in each lora, going with his CG over sprite work, etc.You can get some effect from the Flux lora alone. Just not nearly as well as I'd like. It's recognizably Danganronpa but sometimes 'almost' getting the characters right is more frustrating than if it didn't get them at all. Still, you can try with a strength of 1.0 with the Flux lora and something like this with Sayaka as an example:&lt;lora:Hyper-FLUX.1-dev-8steps-lora:0.125&gt;, &lt;lora:100danganpack-flux:1&gt; , 100danganpack, official_art, komatsuzaki_rui_(style), danganronpa_(series), danganronpa:_trigger_happy_havoc, maizono_sayaka, A detailed anime-style professional illustration which features high quality anime digital art combined with detailed character design. Sayaka Maizono wearing her uniform from Danganronpa 1.I should have an example in the gallery of this Sayaka prompt style and this lora at 0.4 combined with my Sayaka lora for comparison. With this Flux lora alone you'll note that she's inheriting some elements from the Tsumugi sprite and some from the 100 line sprites. Presumably Darumi in particular because of the hair. But the 100 line characters have a certain...high res sheen you could call it. I actually kind of like the impact that has on the standard character lora when this is set at 0.4 strength and the standard character lora is at 1.0. Again, I should have an example of both using Sayaka in the gallery.<h1 id="what-was-the-dataset-curation-methodology">What was the dataset curation methodology?</h1>For the 1.x loras I started out with a base of official artwork from danganronpa and then added in all the higher quality pieces from danbooru that had a crossover between 100 line and danganronpa. Then I scripted out a process to check how many images a specific character had and to then fill in the gaps with quality sorted, tag blacklisted, items from danbooru. For characters who still had gaps I filled them with the best synthetic data that I could. I don't like using synthetic data, but after the amount of gens I've done I have a fair amount I think are viable to beef up characters like Aiko. In total that didn't make for a huge amount of images per character given that I was using civit's trainer and its limit of 1,000 images. My hope was that it wouldn't be a huge issue for modern Illustrious models given that they generally have a pretty good grasp of the major danganronpa characters anyway. Having a nudge in the right direction for well represented characters and a blow to the head to drive in the minor characters better would be enough. I'm more curious how Flux will do. Given that Flux has no knowledge of the franchise but better adaptability to training on smaller datasets.The 2.x versions had the number of 100 line characters reduced and rain code reduced dramatically. I put an emphasis there on crossovers between it and danganronpa. Additional quality control was implemented, synthetic data added to fill out missing elements, more stress was put on multi-character images, and I generally went through the curation process again for a large number of the characters.<h1 id="please-just-forgive-me-already!">Please just forgive me already!</h1>Might as well call me Leon, because when I was putting together the list of major characters for the 1.x lora I forgot Kanon Nakajima. That's fixed in the 2.x.<h1 id="prompting-style">Prompting style?</h1>Underscored, comma seperated, danbooru style tags. Yes, for the Flux one as well. I was thinking more along the lines of using it for Chroma so didn't want to go through the process of re-captioning everything in a narrative style too.IronySo far the strongest recognizable training example for me in the dataset is Tsumugi. The character I already have many, many, many loras for.<h1 id="what-are-the-tags">What are the tags?</h1>The specific tags are going to take a while to get up. I'm in the process of documenting all the danganronpa characters, tags and descriptive, in all the official outfits. But until the just finding an example on danbooru should be the best bet for trying this. Or using any of the images I've posted as a guide.<h1 id="anything-special-about-the-tags-or-clothing">Anything special about the tags or clothing?</h1>Most of the Danganronpa 10th aniversery "prom" outfits were in the training data. Not to a huge extent, they're sparse, but they're in there. Again, specific tag lists to come.<h1 id="what-danganronpa-characters-were-trained-on-for-this-lora">What Danganronpa characters were trained on for this lora?</h1>iruma_miukamukura_izurualter_ego_(danganronpa)ikusaba_mukuronaegi_makotokirigiri_kyokotogami_byakuyafukawa_tokomonokumaishimaru_kiyotakaasahina_aoiyamada_hifumienoshima_junkofujisaki_chihiroowada_mondocelestia_ludenbergkuwata_leonmaizono_sayakaogami_sakuraikusaba_mukurohagakure_yasuhirohinata_hajimekomaeda_nagitonanami_chiakikoizumi_mahirupekoyama_pekokuzuryu_fuyuhikotanaka_gundhamsonia_nevermindsoda_kazuichiowari_akanenidai_nekomarusaionji_hiyokotsumiki_mikanhanamura_teruterutogami_byakuya_(danganronpa_2)naegi_komarudaimon_masaru_(danganronpa)shingetsu_nagisautsugi_kotokokemuri_jatarotowa_monacamitarai_ryotamunakata_kyousukesakakura_juuzoukizakura_kouichigreat_gozuyukizome_chisakimura_seikoando_rurukaizayoi_sounosukegekkogahara_miayaakamatsu_kaedesaihara_shuichiamami_rantaroshirogane_tsumugiyumeno_himikoharukawa_makishinguji_korekiyok1-b0gokuhara_gontamomota_kaitooma_kokichitojo_kirumiyonaga_angiechabashira_tenkoshirogane_tsumugiamami_rantaroakamatsu_kaedeumesawa_aikomatsuda_yasukeotonashi_ryoukoiranami_satsukikurokawa_mikako<h1 id="what-pairings-of-danganronpa-characters-were-trained-on">What pairings of Danganronpa characters were trained on?</h1>shirogane_tsumugi, gokuhara_gontamomota_kaito, harukawa_makinaegi_makoto, kirigiri_kyokonaegi_makoto, maizono_sayakamaizono_sayaka, shirogane_tsumugikuzuryu_fuyuhiko, pekoyama_pekosonia_nevermind, tanaka_gundhamakamatsu_kaede, saihara_shuichisaihara_shuichi, momota_kaitosaihara_shuichi, momota_kaito, harukawa_makik1-b0, iruma_miuchabashira_tenko, yumeno_himikoamami_rantaro, akamatsu_kaedeamami_rantaro, shirogane_tsumugik1-b0, shirogane_tsumugishirogane_tsumugi, enoshima_junkotowa_monaca, enoshima_junkokamukura_izuru, nanami_chiaki<h1 id="what-100-line-characters-were-trained-on">What 100 Line characters were trained on?</h1>sumino_takumikashimiya_karuayakushiji_takemarushizuhara_hirukoamemiya_darumiaotsuki_eitokawana_tsubasamaruko_gakutsukumo_imatsukumo_kakoginzaki_shoumakirifuji_nozomimagadori_kyoshikaomokage_yugamuoosuzuki_kuraramojiro_moko<h1 id="what-rain-code-characters-were-trained-on-(1.x-loras-only)">What Rain Code characters were trained on? (1.x loras only)</h1>shinigami_(rain_code)pucci_lavmin<h1 id="general-in-progress-tag-list:">General in progress tag list:</h1>I'll try to just gradually get tags for everyone in here.Darumi Hemoanima1girl,amemiya_darumi,armband,black_boots,black_bow,black_hair,black_sailor_collar,black_serafuku,black_shirt,black_skirt,blood,blood_from_eyes,blue_hair,blunt_bangs,bodysuit,boots,bow,collarbone,full_body,gradient_hair,hair_bow,highres,holding,holding_knife,holding_weapon,hole_in_chest,hole_on_body,knee_boots,knife,komatsuzaki_rui,last_defense_academy_school_uniform,long_hair,long_sleeves,looking_at_viewer,miniskirt,mole,mole_under_mouth,multicolored_hair,neckerchief,official_art,open_mouth,pleated_skirt,purple_eyes,red_eyes,sailor_collar,school_uniform,score:4,second-party_source,serafuku,shirt,simple_background,skirt,smile,solo,standing,tachi-e,the_hundred_line_-last_defense_academy-,tongue,tongue_out,transparent_background,two_side_up,v-shaped_eyebrows,very_long_hair,weapon,white_armband,white_bodysuit,white_neckerchief

Files with SHA256: 1016cd9e27a100c3f8098662b78c969ddbf67015ca641f0516fb33540c1dfefb