Hi, I’m using the facebook/detr-resnet-50 model for object detection with two different datasets. The first dataset has 2 classes, and the second has 24 classes. I’m using a learning rate of 5e-5 for both. However, I encounter the same error in both cases. For the first dataset, the training runs fi…

Hi there! It looks like the issue might be related to your training process producing invalid bounding boxes (e.g., NaN values) or incorrectly formatted inputs. Here are some suggestions to help resolve the problem: 1. Check Your Dataset: Ensure the bounding boxes in your dataset are in the correc…

Hi, thank you for your response. I’ve implemented all the action points you suggested, but I’m still encountering the error

@risper7 Did you find a solution?

There are many potential causes for this error. So not enough information to determine the cause… Value error in generalized_box_iou for conditional Detr object detection training custom training asserts with “degenerate bboxes” over and over - but bboxes look correct, any debugging insight? #28 A…

ValueError: boxes1 must be in [x0, y0, x1, y1] (corner) format

🤗Transformers

John6666 August 1, 2025, 2:12pm 5

There are many potential causes for this error. So not enough information to determine the cause…

Value error in generalized_box_iou for conditional Detr object detection training
custom training asserts with “degenerate bboxes” over and over - but bboxes look correct, any debugging insight? #28
AssertionError in generalized_box_iou

Topic		Replies	Views
Value error in generalized_box_iou for conditional Detr object detection training 🤗Transformers	5	1234	June 15, 2024
Trainer API object detection 🤗Transformers	2	92	December 29, 2024
ValueError: too many values to unpack (expected 3) using the DETR model 🤗Transformers	1	691	July 10, 2024
Inaccurate bboxes after finetuning DETR Beginners	9	651	October 6, 2024
Fine tuning DETR errors during training Beginners	2	169	July 2, 2025

ValueError: boxes1 must be in [x0, y0, x1, y1] (corner) format

Related topics