Instructions to use microsoft/dit-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use microsoft/dit-base with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoImageProcessor, BeitForMaskedImageModeling processor = AutoImageProcessor.from_pretrained("microsoft/dit-base") model = BeitForMaskedImageModeling.from_pretrained("microsoft/dit-base", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Add upstream DiT MIT license information
#5
by jakewa - opened
The model card currently omits license metadata and the repository has no license file. This adds the MIT license information from the upstream Microsoft UniLM DiT project.
- Add
license: mitto the model-card metadata. - Add the upstream MIT
LICENSE, preserving Microsoft's copyright notice and the full license text (line endings normalized to LF). - Add a model-card License section linking to the local license and its pinned upstream source.
Source: UniLM LICENSE. The upstream README identifies this root license as the project license, and DiT has no separate license file.
Validation: Hugging Face model-card validation passed; license content matches the pinned upstream file after normalizing line endings; git diff --check passed.
julien-c changed pull request status to closed