Method __init__

datasets/clip.py:16–24

(
        self, version="openai/clip-vit-large-patch14", device="cuda", max_length=77
    )

    """Uses the CLIP transformer encoder for text (from Hugging Face)"""

    def __init__(
        self, version="openai/clip-vit-large-patch14", device="cuda", max_length=77
    ):
        super().__init__()
        self.tokenizer = CLIPTokenizer.from_pretrained(version)
        self.transformer = CLIPTextModel.from_pretrained(version)
        self.device = device
        self.max_length = max_length
        self.freeze()

    def freeze(self):
        self.transformer = self.transformer.eval()