georgian-io/Multimodal-Toolkit
A toolkit that extends HuggingFace transformers to combine text embeddings with tabular categorical and numerical features for classification and regression.
Not currently ranked — collecting fresh signals.
star history
This library builds multimodal models on top of pretrained transformer architectures (BERT, ALBERT, etc.) by adding fusion layers that combine transformer outputs with categorical and numerical tabular features. It supports end-to-end training where both the combining module and transformer parameters are fine-tuned for supervised downstream tasks. The toolkit is built on PyTorch and integrates directly with HuggingFace Transformers.
Frequently asked
- What is georgian-io/Multimodal-Toolkit?
- A toolkit that extends HuggingFace transformers to combine text embeddings with tabular categorical and numerical features for classification and regression.
- Is Multimodal-Toolkit open source?
- Yes — georgian-io/Multimodal-Toolkit is open source, released under the Apache-2.0 license.
- What language is Multimodal-Toolkit written in?
- georgian-io/Multimodal-Toolkit is primarily written in Python.
- How popular is Multimodal-Toolkit?
- georgian-io/Multimodal-Toolkit has 618 stars on GitHub.
- Where can I find Multimodal-Toolkit?
- georgian-io/Multimodal-Toolkit is on GitHub at https://github.com/georgian-io/Multimodal-Toolkit.