yoloe
yoloe copied to clipboard
THU-MIG
YOLOE: Real-Time Seeing Anything
使用from ultralytics import YOLOE from ultralytics.models.yolo.yoloe.train_seg import YOLOESegTrainer model = YOLOE("yoloe-11s-seg.pt") # Fine-tune on your segmentation dataset results = model.train( data="coco128-seg.yaml", # Segmentation dataset epochs=80, patience=10, trainer=YOLOESegTrainer, #
作者你好，我在运行val_vp.py的时候遇到有的图片没有标注。 在这一行[preds = model.get_visual_pe(batch["img"], visual=batch["visuals"])](https://github.com/THU-MIG/yoloe/blob/ad5db223cb47fee4de9afe63b790977ae8c03559/ultralytics/models/yolo/yoloe/val.py#L34)打印出来的batch如下： {'im_file': ['/data/lvis/images/train2017/000000000072.jpg'], 'ori_shape': [[...]], 'resized_shape': [[...]], 'ratio_pad': [[...]], 'img': tensor([[[[0.4471, 0.4471, 0.4471, ..., 0.4471, 0.4471, 0.4471], [0.4471, 0.4471, 0.4471, ..., 0.4471, 0.4471, 0.4471], [0.4471, 0.4471,...
作者你好，感谢你的工作。 我在看你的代码的时候，发现训练的过程中好像没有对GT采样，而是把所有GT都作为Visual Prompts。而不像T-Rex2论文里写的，随机在所有GT中挑选N个，N的范围是1~Num_GT。 这样的训练方法会不会导致模型对visual prompt的数量不鲁棒？
I am working on an emergency and disaster response application. Drone footage captures scenes where a target object (e.g., backpack, laptop, bicycle) may appear intermittently. The goal is to detect...
I got the following error: FileNotFoundError: [Errno 2] No such file or directory: 'tools/mobileclip:blt/global_grounding_neg_embeddings.pt' Where I can download it?
为什么在20个类别上的数据集上进行微调过后进行指定类别推理的时候就报错了，中间特征图的大小被固定成训练的类别了这是什么原因牙官方预训练过后的就不会有这个问题。x_cat = torch.cat([xi.view(shape[0], self.nc + self.reg_max * 4, -1) for xi in x], 2) File "/home/xgrids/ZJJ/code/yoloe/ultralytics/nn/modules/head.py", line 486, in x_cat = torch.cat([xi.view(shape[0], self.nc + self.reg_max * 4, -1) for xi...
pip 25.3 will enforce this behaviour change. A possible replacement is to add a pyproject.toml or enable --use- pep517, and use setuptools >= 64. If the resulting installation is not...
Hi! Thank you team for your work! I fine tuned the model according to this colab notebook you provided on a custom dataset with bounding box annotations: [https://colab.research.google.com/drive/1y-r4y_owfFAfyqbqP2t64H7IqjURkKwe?usp=sharing#scrollTo=Ez5mFaoZBL9o] Now i...
h, w = feats[i].shape[2:] if isinstance(feats, list) else (int(feats[i][0]), int(feats[i][1])) ValueError: not enough values to unpack (expected 2, got 0)
Hello, I used the yoloe-11l-seg-det.pt model loaded in yoloe-multi-source-predict-vp train train_vp.py, and the model loading showed AMP: checks passed, but when Starting training reported the above error, I printed the...
