The length of text that the text encoder can handle

[song-wensong]

import torch
from languagebind import LanguageBindVideo, LanguageBindVideoTokenizer, LanguageBindVideoProcessor

pretrained_ckpt = 'LanguageBind/LanguageBind_Video_FT'  # also 'LanguageBind/LanguageBind_Video'
model = LanguageBindVideo.from_pretrained(pretrained_ckpt, cache_dir='./cache_dir')
tokenizer = LanguageBindVideoTokenizer.from_pretrained(pretrained_ckpt, cache_dir='./cache_dir')
video_process = LanguageBindVideoProcessor(model.config, tokenizer)

model.eval()
data = video_process(["your/video.mp4"], ['your text.'], return_tensors='pt')
with torch.no_grad():
    out = model(**data)

print(out.text_embeds @ out.image_embeds.T)

In this code, what is the maximum length of your text? If it exceeds 77, will it be truncated directly?

[LinB203]

Maximum text tokens is 77. If it exceeds 77, it will be truncated directly. This method we just follow CLIP.