OpenGVLab / LAMM

[NeurIPS 2023 Datasets and Benchmarks Track] LAMM: Multi-Modal Large Language Models and Applications as AI Agents
https://openlamm.github.io/
296 stars 16 forks source link

Some images are not contained in the dataset (bamboo_images). #13

Closed cxxgtxy closed 1 year ago

cxxgtxy commented 1 year ago

Hi, Please check the following images are not included (broken) in the bamboo_images. I don't iter the whole dataset. bamboo_images/0da6411bc9c3cf1973fab4a143866cd5cc1d7bd2.tablet_computer_home.jpeg bamboo_images/faa5594fbc9d7ee0d125e58a3cb911092d6be990.lighthouse_person.jpeg

Thanks!

double125 commented 1 year ago

您好, 请检查以下图像是否包含在(损坏的)bamboo_images 中。我不会迭代整个数据集。 竹子图像/0da6411bc9c3cf1973fab4a143866cd5cc1d7bd2.tablet_computer_home.jpeg 竹子图像/faa5594fbc9d7ee0d125e58a3cb911092d6be990.lighthouse_person.jpeg

谢谢!

We have discovered this issue, the information of these two images is broken, we will deal with it as soon as possible and update the bamboo image set. In addition, if you are in a hurry to train our code and model, you can skip these two images directly, which will have little impact on the final model performance.

cxxgtxy commented 1 year ago

Thanks