wxbest001

中文版：分享一下我在Tensor.art训练此Lora的过程。https://tensor.art/models/8067359051855880171.数据集处理图片均来自游戏内截图，裁剪至合适的比例。由于要训练衣服的Lora而不是角色的，所以选择了包含6个角色的共20张图片，其中有一组左侧右侧和背面视角的图片。上传至Tensor.Art, Tensor.Art会自动打标。再给每张图片的标签加上christmas clothes作为触发词。由于自动打标没有识别角色，也加上了角色的触发词，例如shinosawa hiro \(gakuen idolmaster\)。2.参数设置底模用Illustrious, 训练网络模块Lora，触发词用christmas clothes。图片单张重复15次，训练轮数10，每1轮保存一个。Clip Skip 设为2，其他参数按默认。样图提示词 1girl, christmas clothes3.开始训练训练完成后选择样图效果好的Lora即可发布。English Version:Sharing my Lora training process on Tensor.art:https://tensor.art/models/8067359051855880171.Dataset ProcessingThe images are all screenshots from the game, cropped to an appropriate aspect ratio. Since I wanted to train a Lora for the clothes and not the characters, I selected a total of 20 images including 6 characters, with a set of left, right, and back view images.Uploaded to Tensor.Art, which automatically tagged them. Then I added "christmas clothes" to each image tag as a trigger word. Since the automatic tagging didn't recognize the characters, I also added the character trigger words, such as "shinosawa hiro \(gakuen idolmaster\)".2.Parameter SettingsBase model: Illustrious, Network Module Lora, trigger words: "christmas clothes".Image repetitions: 15 per image, training epochs: 10, saving every 1 epoch. Clip Skip: 2, other parameters using default in Tensor.art.Sample image prompt: "1girl, christmas clothes"3.Start TrainingAfter training, select and publish the Lora with the best sample image results.

中文版：使用的AI Tool:Illustrious T2V with Lora by CogVideoXhttps://tensor.art/template/804842224037688462第一步，选择模型和Lora。模型推荐使用Illustrious 系列的模型。Lora在此推荐一个我训练的Lora: 动漫角色画风优化 https://tensor.art/models/807298004735338854无触发词直接使用，可优化动漫角色画风，权重推荐0.8。第二步，输入图片的提示词为了与CogVideoX的设置对应，图片比例固定为1440x960。请输入合适的图片提示词，包括正向提示词和反向提示词。第三步，输入视频的提示词请输入CogVideoX的提示词，包括正向提示词和反向提示词。第四步，生成点击Go即可开始生成， 每次生成会生成一张图片和一个视频。English Version:Used AI Tool:Illustrious T2V with Lora by CogVideoXhttps://tensor.art/template/804842224037688462Step 1: Select the Model and LoraRecommended to use models from the Illustrious series.For Lora, I recommend one that I trained: Anime Character Art Style Optimization https://tensor.art/models/807298004735338854Directly use this Lora with out trigger word, it can optimize the anime style of character. Recommended strength:0.8.Step 2: Input Image PromptsTo match CogVideoX settings, the image ratio is fixed at 1440x960. Please enter appropriate image prompts, including both positive and negative prompts.Step 3: Input Video PromptsPlease enter CogVideoX prompts, including both positive and negative prompts.Step 4: GenerateClick Go to start generating. Each generation will produce one image and one video.