Qwen3_5VideoConverter

[source]

Qwen3_5VideoConverter class

keras_hub.layers.Qwen3_5VideoConverter(
    patch_size=16,
    temporal_patch_size=2,
    spatial_merge_size=2,
    min_pixels=65536,
    max_pixels=16777216,
    interpolation="bicubic",
    antialias=True,
    **kwargs
)

Video pre-processor for Qwen3.5.

Converts videos to the patch tensor format expected by Qwen3_5VisionEncoder and also returns grid_thw metadata.

Arguments

• patch_size: int. Spatial size of each patch in pixels. Default 16.
• temporal_patch_size: int. Frames grouped per temporal patch. Default 2.
• spatial_merge_size: int. Spatial merge downsampling factor. Default 2.
• min_pixels: int. Minimum pixel budget for the resized frames. Frames smaller than this will be upscaled. Default 65536.
• max_pixels: int. Maximum pixel budget. Frames larger than this will be downscaled. Default 16777216 (= 4096×4096).
• interpolation: str. Interpolation method for resizing. Defaults to "bicubic".
• antialias: bool. Whether to apply antialiasing when resizing. Defaults to True.
• scale: float or list of floats. Per-channel scale for normalisation.
• offset: float or list of floats. Per-channel offset for normalisation.

[source]

from_preset method

Qwen3_5VideoConverter.from_preset(preset, **kwargs)

Instantiate a keras_hub.layers.VideoConverter from a model preset.

A preset is a directory of configs, weights and other file assets used to save and load a pre-trained model. The preset can be passed as one of:

1. a built-in preset identifier like 'gemma4_2b_it'
2. a Kaggle Models handle like 'kaggle://user/gemma4/keras/gemma4_2b_it'
3. a Hugging Face handle like 'hf://google/gemma-4-2b-it'
4. a path to a local preset directory like './gemma4_2b_it'

You can run cls.presets.keys() to list all built-in presets available on the class.

This constructor can be called in one of two ways. Either from the base class like keras_hub.models.VideoConverter.from_preset(), or from a model class like keras_hub.models.Gemma4VideoConverter.from_preset(). If calling from the base class, the subclass of the returning object will be inferred from the config in the preset directory.

Arguments

• preset: string. A built-in preset identifier, a Kaggle Models handle, a Hugging Face handle, or a path to a local directory.

Examples

# Load a video converter from a preset.
converter = keras_hub.layers.VideoConverter.from_preset(
    "hf://google/gemma-4-2b-it"
)