Weihua Chen

Did you use the wrong version of Detectron2? The COCOEvaluator in Detectron2=0.3 is as below: 

I've changed my Detectron2 from v0.3 to v0.5. The code can be well run. But the performance is much lower than the reported, as below (city to foggy). It's hard...

Here is my config.yml: CUDNN_BENCHMARK: false DATALOADER: ASPECT_RATIO_GROUPING: true FILTER_EMPTY_ANNOTATIONS: true NUM_WORKERS: 4 RANDOM_DATA_SEED: 0 RANDOM_DATA_SEED_PATH: dataseed/COCO_supervision.txt REPEAT_THRESHOLD: 0.0 SAMPLER_TRAIN: TrainingSampler SUP_PERCENT: 100.0 DATASETS: CROSS_DATASET: true PRECOMPUTED_PROPOSAL_TOPK_TEST: 1000 PRECOMPUTED_PROPOSAL_TOPK_TRAIN: 2000...

By the way, when I run the code on city to foggy, the dataset information is listed as below. Do it the same with yours?  When I run the...

> @cwhgn check if comment this line works for you for the clipart? https://github.com/facebookresearch/adaptive_teacher/blob/main/adapteacher/data/dataset_mapper.py#L83 Thanks, it works. By the way, can you give me some help on the low performance...

> @cwhgn Are you looking at AP50 instead of AP? For AP50 of each class, please find it in the tensorboard. Thanks, it's a great help for me! Another question,...

Groundtruth.txt是提供的检测的groundtruth，不是跟踪的。

是的，windows下用vs2008运行，具体环境请参考Code Prerequisites中的要求~

你好，训练好的model可以在[SOLIDER主页](https://github.com/tinyvision/SOLIDER)的Models里找到。

SOLIDER提供的pretrained model是作为下游任务（比如PedestrianDetection）的初始化预训练模型用的，其作用类似于Swin。下游任务在使用了预训练模型之后，也是需要专门训练的，并不能用预训练模型直接inference。