wangchen1801

Hello, here is the completed code, you can try this out. Feel free to ask if you have any other questions. ```python # out = query_feature + self.gamma * out...

Sorry for the late reply, please try to test images in different dataset: configs/__base__/datasets/nway_kshot/few_shot_voc_ms.py ```python test=dict( type='FewShotVOCDataset', # ann_cfg=[ # dict( # type='ann_file', # ann_file=data_root + 'VOC2007/ImageSets/Main/test.txt'), # ], #...

Hello, I checked the previous code and you are right. We need to add "img_metas[0]['img'] = img" at: fpd/fpd_detector.py ```python def simple_test(self, img: Tensor, img_metas: List[Dict], proposals: Optional[List[Tensor]] = None,...

Hello yefeng, transforms.py contains some image preprocessing methods. See the file below for more functions. ```shell mmfewshot/mmfewshot/detection/datasets/pipelines/transforms.py ``` For example, CropResizeInstance defines how to obtain a fixed-size image patch (224,...

Hello, transforms.py is only used while training on COCO dataset. Your understanding is correct, we don't need to modify the code. Best~

Thank you for your interest in this work! To fit a custom dataset, we should change the following configs. Please ensure the custom dataset has been configured properly. * configs/fpd/voc/split1/fpd_r101_c4_2xb4_voc-split1_10shot-fine-tuning.py...

Hello, this operation can scale up attention values. With the subsequent softmax function, it leads to a sharper focus on the key elements that the attention mechanism is trying to...

In our experiments, feature queries are class-agnostic initially. It is reasonable to assign a class embedding to each category for discrimination. However, in the final experiment, each category has its...

Hello, it seems that your mmfewshot didn't install correctly. Please make sure to follow the steps below: ```python # install mmfewshot pip install git+https://github.com/open-mmlab/mmfewshot.git # or manually download the code,...

Please try this: ```python pip install yapf==0.40.1 ``` it can solve the problem in my evironment~