custom video error

[pradeepkc11]

Hi while running customized video of my file i am getting this error

pred, identity_attentions, aggregated_attentions, identities, frames_per_identity = predict(opt.video_path, clustered_faces, config, opt)
File "/home/pradeep/MINTIME-Multi-Identity-size-iNvariant-TIMEsformer-for-Video-Deepfake-Detection/predict.py", line 393, in predict
videos, size_embeddings, mask, identities_mask, positions, tokens_per_identity = generate_masks(video_path, identities, discarded_faces, config["model"]["num-frames"], config["model"]["image-size"], config["model"]["num-patches"])
File "/home/pradeep/MINTIME-Multi-Identity-size-iNvariant-TIMEsformer-for-Video-Deepfake-Detection/predict.py", line 280, in generate_masks
identity_images = np.asarray(identity_images)[idx]
ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 2 dimensions. The detected shape was (20, 3) + inhomogeneous part.

[davide-coccomini]

Can you check that all the frames you are passing have the same size?