Text2video zero refinements (#3733)
* fix docs typos. add frame_ids argument to text2video-zero pipeline call
* make style && make quality
* add support of pytorch 2.0 scaled_dot_product_attention for CrossFrameAttnProcessor
* add chunk-by-chunk processing to text2video-zero docs
* make style && make quality
* Update docs/source/en/api/pipelines/text_to_video_zero.mdx
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
