-
Hi, I have created a WandB report that is attached to the issue that outlines the exact model and loss logs of the model.
[WB_report.pdf](https://github.com/open-mmlab/mmdetection/files/7270223/WB_r…
-
Hi, thank you for this amazing work !! I was trying out your torchgeo_demo.ipynb on my local machine. I only changed the 'max_epochs' parameter and ran the rest of the code blocks as it is.
I let t…
-
when applying the code below the result is equal to Tag2Text not equal to the RAM result as you refer to in the repo.
export CUDA_VISIBLE_DEVICES=0
python automatic_label_ram_demo.py \
--config G…
-
Dear author,
I notice that the repo recommends using apex mixed-precision for fine-tuning.
Then, how about learning from scratch on ImageNet-1k (should I also open the Apex mixed-precision traini…
-
Hi, thanks to your effort on this fantastic work!
I encountered some difficulties while reproducing the experiment.
I am planning to train Swin-B on the Cityscapes dataset and finetune it on coco a…
-
- [ ] review papers
- [ ] check official implementations
- [ ] choose one of them
- [ ] start implementing
mnn59 updated
2 years ago
-
https://github.com/SorinaBoz/OOPHomeworkInterfaces/blob/205cbe1f5e3c917671be9a7e27cfbd53ce087724/Swimmer.java#L1
In cerinta scria sa se implementeze 2 interfete Swim si Breath.
Din moment ce tu ai…
-
Hi Hu Cao:
Thanks for your great work and I am very interested in swin-unet. I use this git repo to train my own data but the predicted segmentation mask seems like composed of image patches and the …
-
可以提供一下训练best_31.9_swin_tiny_coco_308000.pth中使用tensorboard记录的各种loss和mAP吗？想与自己的结果比较和分析一下，十分感谢！
-
conduct python tools/test.py configs/mask2former/mask2former_swin-l.py work_dirs/mask2former_swin-l/best_mIoU_iter_100000.pth --cfg-options test_evaluator.type=IoUMetric
01/08 19:36:52 - mmengine -…