zf_qwen3_vl_process / processor_config.json
TYTTYTTYT's picture
adjust the default video processing parameters for long videos, add focus_size parameter
2cdaef1 verified
{
"auto_map": {
"AutoProcessor": "processing_qwen3_vl.ZFQwen3VLProcessor"
},
"image_processor": {
"auto_map": {
"AutoImageProcessor": "image_processing_qwen2_vl_fast.ZFQwen2VLImageProcessorFast",
"AutoProcessor": "processing_qwen3_vl.ZFQwen3VLProcessor"
},
"data_format": "channels_first",
"do_convert_rgb": true,
"do_normalize": true,
"do_rescale": true,
"do_resize": true,
"focus_size": 2,
"image_mean": [
0.5,
0.5,
0.5
],
"image_processor_type": "ZFQwen2VLImageProcessorFast",
"image_std": [
0.5,
0.5,
0.5
],
"merge_size": 2,
"patch_size": 16,
"resample": 3,
"rescale_factor": 0.00392156862745098,
"size": {
"longest_edge": 16777216,
"shortest_edge": 65536
},
"temporal_patch_size": 2
},
"processor_class": "ZFQwen3VLProcessor",
"video_processor": {
"auto_map": {
"AutoProcessor": "processing_qwen3_vl.ZFQwen3VLProcessor",
"AutoVideoProcessor": "video_processing_qwen3_vl.ZFQwen3VLVideoProcessor"
},
"data_format": "channels_first",
"default_to_square": true,
"do_convert_rgb": true,
"do_normalize": true,
"do_rescale": true,
"do_resize": true,
"do_sample_frames": true,
"focus_size": 2,
"fps": 1,
"image_mean": [
0.5,
0.5,
0.5
],
"image_std": [
0.5,
0.5,
0.5
],
"max_frames": 3601,
"merge_size": 2,
"min_frames": 4,
"patch_size": 16,
"resample": 3,
"rescale_factor": 0.00392156862745098,
"return_metadata": false,
"size": {
"longest_edge": 235929601,
"shortest_edge": 4096
},
"temporal_patch_size": 2,
"video_processor_type": "ZFQwen3VLVideoProcessor"
}
}