hoya-cho
hoya-cho
hello. I want to do early stopping based on the validation loss. I have seen the questions and answers through the following link. https://github.com/open-mmlab/mmcv/issues/1701 so you referenced https://github.com/open-mmlab/mmdetection/blob/master/mmdet/core/evaluation/eval_hooks.py I'm looking...
I am doing early stopping based on validation loss value or mAP value on multi-GPU. I receive mAP from rank 0 and store the best ckpt, at this time, I...
In train_detector(), why "TypeError: is unsupported for json dump " error occurs? in train_detector(model, dataset, cfg, distributed, validate, timestamp, meta, runner, progress_callback) 258 elif cfg.load_from: 259 runner.load_checkpoint(cfg.load_from) --> 260 runner.run(data_loaders,...
The format of crowdhuman data is known as the odgt format. Are you using the crowdhuman data format converted to coco format? Or do you ignore coordinate values or attribute...
hello. Is there a way to know the truncation of objects at the edges of the image during amodal completion inference? Or is there a way to apply it? thank...
### Prerequisite - [X] I have searched [the existing and past issues](https://github.com/open-mmlab/mmyolo/issues) but cannot get the expected help. - [X] I have read the [FAQ documentation](https://mmyolo.readthedocs.io/en/latest/faq.html) but cannot get the...
How to infer image data other than dancetrack? And can the pretrained weights be inferred using the pretrained model at https://github.com/fundamentalvision/Deformable-DETR? Lastly, why do we need det_db_motr.json in simple_inference.sh when...
When the score of the result value detected by the detection model is around 0.3 or 0.4, and the corresponding bbox is captured well even though the score is low,...
How can I set multi gpu environment in my program like multi gpu using torch.distributed.launch in dist_train.sh? Even if the master port and master address are set in os.environ, local_rank...
I am currently using strongsort. The detector used at that time is yolox, but I want to change the detector to codetr. To do so, I need to create a...