Method generate_once

evaluation/vqlm_demo/inference.py:206–207

(input_images, n_new_frames, temperature=1.0, top_p=1.0)

    @staticmethod
    def generate_once(input_images, n_new_frames, temperature=1.0, top_p=1.0):
        return _current_process_backend.generate_once(input_images, n_new_frames, temperature, top_p)

    @staticmethod
    def compute_perplexity_once(input_images, target_images):