Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

how to train it on custom dataset #27

Open
muhammadhammadjamil opened this issue Nov 2, 2023 · 2 comments
Open

how to train it on custom dataset #27

muhammadhammadjamil opened this issue Nov 2, 2023 · 2 comments

Comments

@muhammadhammadjamil
Copy link

@yjh0410 can you provide guildline how to train YOWOv2 on my own dataset here is alteration i make on dataset config file
'ava_v2.2': {
# dataset
'frames_dir': '/home/ubuntu/Furqan/new_dataset/20231004-activity-all-annotated-tasks/',
'frame_list': '/home/ubuntu/hammad/YOWOv2/config/frame/',
'annotation_dir': '/home/ubuntu/hammad/new_annotations/',
'train_gt_box_list': '/home/ubuntu/hammad/new_annotations/20231005_ke_activity_train_avastyle_17.csv',
'val_gt_box_list': '/home/ubuntu/hammad/new_annotations/20231005_ke_activity_val_avastyle_17.csv',
'train_exclusion_file': '/home/ubuntu/hammad/new_annotations/ava_train_excluded_timestamps_v2.2 (1).csv',
'val_exclusion_file': '/home/ubuntu/hammad/new_annotations/ava_val_excluded_timestamps_v2.2.csv',
'labelmap_file': '/home/ubuntu/hammad/YOWOv2/labels.pbtxt', # 'ava_v2.2/ava_action_list_v2.2.pbtxt',
'class_ratio_file': None,
'backup_dir': 'results/',

    **however it give me this error** 

Model Config: YOWO_V2_LARGE
Finished loading image paths from: /home/ubuntu/hammad/YOWOv2/config/frame/train.csv
Traceback (most recent call last):
File "train.py", line 330, in
train()
File "train.py", line 156, in train
dataset, evaluator, num_classes = build_dataset(d_cfg, args, is_train=True)
File "/home/ubuntu/hammad/YOWOv2/utils/misc.py", line 67, in build_dataset
dataset = AVA_Dataset(
File "/home/ubuntu/hammad/YOWOv2/dataset/ava.py", line 49, in init
self._load_data()
File "/home/ubuntu/hammad/YOWOv2/dataset/ava.py", line 65, in _load_data
boxes_and_labels = ava_helper.load_boxes_and_labels(
File "/home/ubuntu/hammad/YOWOv2/dataset/ava_helper.py", line 134, in load_boxes_and_labels
if box_key not in all_boxes[video_name][frame_sec]:
KeyError: 0

@xingjinderhuoche
Copy link

can you provide the source dataset ucf24 ?I can not download that. Here is my email [email protected]. Thank you.

@T-wow
Copy link

T-wow commented Jul 1, 2024

你好,请问,请问为什么进行DDP训练损失异常的大,但是进行单卡训练没有这种情况。谢谢解答

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants