hwjiang1510 / VQLoC

(NeurIPS 2023) Open-set visual object query search & localization in long-form videos

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

KeyError: 'response_track_valid_range'

fb-reps opened this issue · comments

Hello,
This is a great work.
But I can't find key 'response_track_valid_range' in any json file.

I see that you seem to generate train_anno_new.json and val_anno_new.json.
but I can not get them when I follow the readme step.


Caught KeyError in DataLoader worker process 0.
Original Traceback (most recent call last):
File "/home/UNT/bf0191/anaconda3/envs/vqloc2/lib/python3.8/site-packages/torch/utils/data/_utils/worker.py", line 302, in _worker_loop
data = fetcher.fetch(index)
File "/home/UNT/bf0191/anaconda3/envs/vqloc2/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py", line 49, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/UNT/bf0191/anaconda3/envs/vqloc2/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py", line 49, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/mnt/DATA-1/bingfanprojects/VQLoC/dataset/base_dataset.py", line 363, in getitem
clip, clip_idxs, before_query = self.clip_reader(clip_path,
File "/mnt/DATA-1/bingfanprojects/VQLoC/dataset/base_dataset.py", line 477, in read_frames_decord_balance
frame_idxs = sample_frames_balance(num_frames, query_frame, frame_interval, sample, sampling) # downsampled fps idxs, used to get bbox annotation
File "/mnt/DATA-1/bingfanprojects/VQLoC/dataset/base_dataset.py", line 428, in sample_frames_balance
anno_valid_idx_range = sample["response_track_valid_range"]
KeyError: 'response_track_valid_range'

I can find the key exists in train_anno_new.json
Are you parsing the data correctly?

image

Great.
how can i get the train_anno_new.json.
I follow the vqloc readme file to use vq2d json process, I get val_annot.json and train_annot.json, but not train_anno_new.json

Thank you. I got it.