AI绘画之训练自己的Lora模型

不曾逗留

感谢分享666

mikuyaya

很棒，学习了

醉卧二两仙

prepare tokenizer
update token length: 225
Using DreamBooth method.
prepare images.
0 train images with repeating.
0 reg images.
no regularization images / 正則化画像が見つかりませんでした
[Dataset 0]
  batch_size: 1
  resolution: (512, 512)
  enable_bucket: True
  min_bucket_reso: 256
  max_bucket_reso: 1024
  bucket_reso_steps: 64
  bucket_no_upscale: False


[Dataset 0]
loading image sizes.
0it [00:00, ?it/s]
make buckets
number of images (including repeats) / 各bucketの画像枚数（繰り返し回数を含む）
E:\新建文件夹\lora训练\lora-scripts\venv\lib\site-packages\numpy\core\fromnumeric.py:3504: RuntimeWarning: Mean of empty slice.
  return _methods._mean(a, axis=axis, dtype=dtype,
E:\新建文件夹\lora训练\lora-scripts\venv\lib\site-packages\numpy\core\_methods.py:129: RuntimeWarning: invalid value encountered in scalar divide
  ret = ret.dtype.type(ret / rcount)
mean ar error (without repeats): nan
No data found. Please verify arguments (train_data_dir must be the parent of folders with images) / 画像がありません。引数指定を確認してください（train_data_dirには画像があるフォルダではなく、画像があるフォルダの親フォルダを指定する必要があります）
Train finished

图片素材加载不上，检查素材路径没问题，。。。。。

215407819@qq.co

这个专业

lovemmm520

插眼收藏

marioking

请问楼主，无法预训练图片是什么地方出错了呢？重装系统前没有这个问题

*** Error completing request
*** Arguments: ('task(rnm9khqu490nerr)', 'E:\\lora训练环境\\lora-scripts\\train\\j_in\\x2', 'E:\\lora训练环境\\lora-scripts\\train\\j', 512, 512, 'ignore', False, False, False, False, True, 0.5, 0.2, True, 0.9, 0.15, 0.5, False, False, 384, 768, 4096, 409600, 'Maximize area', 0.1) {}
    Traceback (most recent call last):
      File "E:\novelai-webui-aki-v2\modules\call_queue.py", line 57, in f
        res = list(func(*args, **kwargs))
      File "E:\novelai-webui-aki-v2\modules\call_queue.py", line 36, in f
        res = func(*args, **kwargs)
      File "E:\novelai-webui-aki-v2\modules\textual_inversion\ui.py", line 19, in preprocess
        modules.textual_inversion.preprocess.preprocess(*args)
      File "E:\novelai-webui-aki-v2\modules\textual_inversion\preprocess.py", line 18, in preprocess
        preprocess_work(process_src, process_dst, process_width, process_height, preprocess_txt_action, process_keep_original_size, process_flip, process_split, process_caption, process_caption_deepbooru, split_threshold, overlap_ratio, process_focal_crop, process_focal_crop_face_weight, process_focal_crop_entropy_weight, process_focal_crop_edges_weight, process_focal_crop_debug, process_multicrop, process_multicrop_mindim, process_multicrop_maxdim, process_multicrop_minarea, process_multicrop_maxarea, process_multicrop_objective, process_multicrop_threshold)
      File "E:\novelai-webui-aki-v2\modules\textual_inversion\preprocess.py", line 212, in preprocess_work
        for focal in autocrop.crop_image(img, autocrop_settings):
      File "E:\novelai-webui-aki-v2\modules\textual_inversion\autocrop.py", line 32, in crop_image
        focus = focal_point(im_debug, settings)
      File "E:\novelai-webui-aki-v2\modules\textual_inversion\autocrop.py", line 75, in focal_point
        face_points = image_face_points(im, settings) if settings.face_points_weight > 0 else []
      File "E:\novelai-webui-aki-v2\modules\textual_inversion\autocrop.py", line 150, in image_face_points
        faces = detector.detect(np.array(im))
    cv2.error: OpenCV(4.8.0) D:\a\opencv-python\opencv-python\opencv\modules\dnn\src\net_impl.cpp:279: error: (-204:Requested object was not found) Layer with requested id=-1 not found in function 'cv::dnn::dnn4_v20230620::Net::Impl::getLayerData'

哈哈哈19908

感谢

xhy1983

非常好的东西

haohmaru

要学的东西好多