KaihuaTang / Scene-Graph-Benchmark.pytorch

A new codebase for popular Scene Graph Generation methods (2020). Visualization & Scene Graph Extraction on custom images/datasets are provided. It's also a PyTorch implementation of paper “Unbiased Scene Graph Generation from Biased Training CVPR 2020”
MIT License
1.03k stars 228 forks source link

VGDataset' object has no attribute 'gt_classes' #103

Open mukesh-2538 opened 3 years ago

mukesh-2538 commented 3 years ago

❓ Questions and Help

mukesh-2538 commented 3 years ago

@KaihuaTang while running the command for custom dataset , i printed out the custom eval value For the First time it is True and the second time it is false. why is it so? Why is init called twice?

During this second call i get the error List index out of range of this line

self.filenames, self.img_info = load_image_filenames(img_dir, image_file) self.filenames = [self.filenames[i] for i in np.where(self.split_mask)[0]] self.img_info = [self.img_info[i] for i in np.where(self.split_mask)[0]] 2020-11-23 10:18:19,069 maskrcnn_benchmark.data.build INFO: ---------------------------------------------------------------------------------------------------- 2020-11-23 10:18:19,069 maskrcnn_benchmark.data.build INFO: get dataset statistics... hello there the custom eval value is : True /content/images hello there the custom eval value is : False

Fred199683 commented 3 years ago

hello! I am also working on testing on my own custom images. Do u have WeChat or other social network, I think we can communicate on that rather than GitHub...

Lyon52222 commented 3 years ago

@mukesh-2538 Hello, I also met this problem when I testing on my custom dataset, do you solved this problem? Thanks