| { | |
| "add_start_end_special_tokens": true, | |
| "auto_map": { | |
| "AutoImageProcessor": "image_processor.MiniMaxM3VLImageProcessor", | |
| "AutoProcessor": "processing_minimax.MiniMaxVLProcessor", | |
| "AutoVideoProcessor": "video_processor.MiniMaxM3VLVideoProcessor" | |
| }, | |
| "do_convert_rgb": true, | |
| "do_normalize": true, | |
| "do_rescale": true, | |
| "do_resize": true, | |
| "image_mean": [ | |
| 0.48145466, | |
| 0.4578275, | |
| 0.40821073 | |
| ], | |
| "image_processor_type": "MiniMaxM3VLImageProcessor", | |
| "image_std": [ | |
| 0.26862954, | |
| 0.26130258, | |
| 0.27577711 | |
| ], | |
| "img_token_compression_config": { | |
| "image_token_compression_method": "patch_merge", | |
| "image_token_compression_threshold": 1.1, | |
| "max_image_resolution": 1008, | |
| "spatial_merge_size": 2, | |
| "temporal_patch_size": 2 | |
| }, | |
| "merge_size": 2, | |
| "patch_size": 14, | |
| "process_image_mode": "dynamic_res", | |
| "resample": 3, | |
| "rescale_factor": 0.00392156862745098, | |
| "size": { | |
| "longest_edge": 451584, | |
| "shortest_edge": 3136 | |
| }, | |
| "temporal_patch_size": 2 | |
| } | |