Tony Wu
In the unit 5 of the audio course, the following code is used: ```python class DataCollatorSpeechSeq2SeqWithPadding: processor: Any def __call__( self, features: List[Dict[str, Union[List[int], torch.Tensor]]] ) -> Dict[str, torch.Tensor]: #...
## Description ### Added - Restore, refactor, and improve the `interpretability` module for generating similarity maps ## Tests The following figures were obtained using the code from https://github.com/tonywu71/colpali-cookbooks/pull/5. ### ColPali...
## Description Implement and train ColPali2. The most noticeable feature of ColPali2 is to being able to output both single-vector (dense) and multi-vector (ColBERT-like) embeddings. Follow-up of https://github.com/illuin-tech/colpali/pull/22.
## Description The `interpretability` module has been been moved, refactored and improved in https://github.com/illuin-tech/colpali/pull/96. ## Features ### Deprecated - Deprecate the `interpretability` module
## Description The interpretability module was moved from [`vidore-benchmark`](https://github.com/illuin-tech/vidore-benchmark) to [`colpali`](https://github.com/illuin-tech/colpali) and refactored in https://github.com/illuin-tech/colpali/pull/96. This PR adapts the similarity map notebook to take these changes into account. ## Features...
# What does this PR do? Add [ColPali](https://doi.org/10.48550/arXiv.2407.01449) support in 🤗 `transformers`. ## Who can review? @yonigozlan 😉 ## Additional details - This PR uses the new [Modular 🤗 transformers](https://huggingface.co/docs/transformers/main/en/modular_transformers#modular-transformers)...
## Description At the following line, the `relavant_docs` dictionary is overridden instead of getting updated. This causes `relavant_docs` to only contain the last seen query. This is problematic in datasets...
## Description Add support for the ColSmol model family. Reference thread: https://x.com/tonywu_71/status/1882434084657590522. ## Changes ### Added - Add support for ColSmol model ### Tests - Add test for ColSmol model