google/tipsv2-g14

Use relative imports for sibling modules (fixes local loading, save_pretrained, pickling)

Support sentencepiece 0.2.2 in text tokenizer

Add arXiv tag to link the paper

Add arXiv paper link to citation

Update files for transformers integration (#1)

Fix broken image URL and add missing .cpu() in code snippets

Remove zero-shot segmentation snippet, link to GitHub for advanced use cases

PCA visualization: add whitening and sigmoid

Add navigation table linking all variants and DPT heads

Switch to bus image with matching classes

Improve code examples: better classes, descriptive comments

Use cat photo, add print statements to code examples

Fix meta device crash: use pure Python for drop_path_rate

Fix zero-shot segmentation section, use public example image

Update README

Align image_encoder.py drop_path_rate computation with upstream

Upload README.md with huggingface_hub

Upload model.safetensors with huggingface_hub

Upload README.md with huggingface_hub

Upload config.json with huggingface_hub

Upload tokenizer.model with huggingface_hub

Upload text_encoder.py with huggingface_hub

Upload image_encoder.py with huggingface_hub

Upload modeling_tips.py with huggingface_hub

Upload configuration_tips.py with huggingface_hub

initial commit