wbw520
wbw520
I think there is a print for matching matirx in line MTUNet/model/FSL.py line 141. When you run vis_fsl.py, "args.vis" is set to True and there should be a printed numpy...
It is drawn by combining the numpy matrix and attention images with the drawing software (not drawn by the code).
The split of cifarfs is based on the code from https://github.com/bertinetto/r2d2. It uses the CSV file which is also utilized in the data loader for my code. If you add...
This is a Pytorch version problem and there are two solutions for this. 1. In model/scouter/scouter_attention.py line 19 add torch.abs() outside the torch.randn(). 2. Using torch 1.6.0 and torchvision 0.7.0
Hi, I think this is a bug by multi-gpu setting. I run the code only with multi-gpu. If you pre-train the model in one gpu with "--multi_gpu False", one solution...
The test.py is actually used for evaluate mAP. If you want to try to visulaize your own image or video, please used the tools/factory.py. I have updated a new version...
It is totally ok.
You can download the weight in the following link https://drive.google.com/file/d/1R2yPUdGUudr3ADTNdsrUgeXye0QGfdwN/view?usp=sharing
I forgot to update "saved_model" folder. It need to be created under the workspace for saving the weight of the model.
Thanks for the question. We calculate the similarity of concept (y) and corresponding class label (label) among a batch data. Thus, it is a matrix similarity score. It will force...