lucrbrtv/doom-e1-internet-gameplay

--- dataset_info: features: - name: video_idx dtype: int32 - name: frame_idx dtype: int32 - name: frame dtype: image - name: action list: float32 splits: - name: train num_bytes: 6516902919 num_examples: 392370 download_size: 6515480616 dataset_size: 6516902919 configs: - config_name: default data_files: - split: train path: data/train-* license: unlicense size_categories: - 100K<n<1M --- I've used the Inverse Dynamic Model, I've previously trained on manually recorded gameplay, on pure gameplay YouTube videos. This dataset is in public domain, use it however you want.