IndexError: index 3327 is out of bounds for axis 0 with size 3327

[mrhanCH]

[INFO] Trainer: ngp | 2023-09-27_19-40-14 | cuda | fp16 | trial_t2/ [INFO] #parameters: 683509 [INFO] Loading latest checkpoint ... [INFO] Latest checkpoint is trial_t2/checkpoints/ngp_ep0042.pth [INFO] loaded model. [INFO] load at epoch 42, global step 127134 [INFO] loaded optimizer. [INFO] loaded scheduler. [INFO] loaded scaler. [INFO] load 100 val frames. [INFO] load aud_features: torch.Size([3325, 1024, 2]) Loading val data: 100%|███████████████████████████████████████████████████████████████████████████████████████| 100/100 [00:00<00:00, 3601.65it/s] [INFO] eye_area: 0.0 - 1.0 [INFO] max_epoch = 42 [INFO] load 303 test frames. [INFO] load aud_features: torch.Size([3325, 1024, 2]) Loading test data: 99%|█████████████████████████████████████████████████████████████████████████████████████▏| 300/303 [00:00<00:00, 3633.58it/s] Traceback (most recent call last): File "/root/autodl-tmp/ER-NeRF/main.py", line 257, in test_loader = NeRFDataset(opt, device=device, type='test').dataloader() File "/root/autodl-tmp/ER-NeRF/nerf_triplane/provider.py", line 500, in init area = au_blink[f['img_id']] IndexError: index 3327 is out of bounds for axis 0 with size 3327

你好，感谢你们的工作和分享，我在使用一个2分钟多的视频进行训练时，当进行测试集加载时就报上面的错误，请问这是哪里的问题，感谢！

[Fictionarry]

检查一下au.csv和ffmpeg抽取的帧数是否相等，问题可能跟视频的编码或者帧率有关