cherche - 2.2.1

Add sentence-transformers as requirements.

- Python
Published by raphaelsty over 1 year ago

cherche - 2.2.0

Update requirements: bump LeNLP version.

- Python
Published by raphaelsty over 1 year ago

cherche - 2.1.1

Update LeNLP version which could raise error on older linux version.

- Python
Published by raphaelsty over 1 year ago

cherche - 2.1.0

Cherche 2.1.0 is available, it's been a long time without updates 🥳

• Updated default TfIdf retriever with vectorizer from LeNLP which is written in Rust and much faster.
• Added BM25 retriever using LeNLP BM25Vectorizer with is written in Rust and very fast.

- Python
Published by raphaelsty over 1 year ago

cherche - 2.0.6

Assert top-k parameter in init method of Lunr is used.

- Python
Published by raphaelsty over 2 years ago

cherche - 2.0.5

Accelerate TfIdf in batches and enables to update TfIdf with new documents using the add method.

- Python
Published by raphaelsty over 2 years ago

cherche - 2.0.4

Add method .to_csr() and .to_csc() to TfIdf in order to speed-up the computation with large batch.

- Python
Published by raphaelsty over 2 years ago

cherche - 2.0.3

Disable tqdm bar of retrievers on demand using tqdm_bar parameter of __call__ methods and add methods.

- Python
Published by raphaelsty over 2 years ago

cherche - 2.0.2

Speed up tf-idf by transposing the sparse matrix of documents in the retriever.add method instead of transposing the sparse matrix when querying.

- Python
Published by raphaelsty over 2 years ago

cherche - 2.0.1

Requirements update of Numpy and Scikit-learn dependencies version.

- Python
Published by raphaelsty over 2 years ago

cherche - 2.0.0

Excited to announce the release of Cherche 2.0, a comprehensive open-source search engine toolkit for Python. This new version comes with a host of new features and improvements, including:

• Batch-computation
• Optimization
• Progress bars
• Cross-Encoders compatibility
• Focus on retrievers, rankers and indexes compatible with Python.
• Requirements are lighter modular

- Python
Published by raphaelsty almost 3 years ago

cherche - 1.0.1

Removed the dependency with grpcio that can cause problems during installation.

- Python
Published by raphaelsty over 3 years ago

cherche - 1.0.0

What's Changed

Here is an essential update for Cherche! 🥳

• Added compatibility with two new open-source retrievers: Meilisearch and TypeSense.
• Compatibility with the Milvus index to use the retriever.Encoder and retriever.DPR models on massive corpora.
• Compatibility with the Milvus index to store ranker embeddings in a database rather than in memory.
• Progress bar when pre-computing embeddings by Encoder, DPR retrievers and Encoder, DPR rankers.
• The path parameter is no longer used.
• All pipelines (voting, intersection, concatenation) produce a similarity score. To do so, the pipeline object applies a softmax to normalize the scores, thus allowing us to "compare" the scores of two distinct models.
• Integration of collaborative filtering models via adding a Recommend retriever and a Recommend ranker (indexation via Faiss and compatible with Milvus) to consider users' preferences in the search.

Cherche is now fully compatible with large-scale corpora and deeply integrates collaborative filtering. Updates retains the previous API and is compatible with previous versions.

- Python
Published by raphaelsty over 3 years ago

cherche - 0.1.0

Added compatibility with the ONNX environment and quantization to significantly speed up sentence transformers and question answering models. 🏎

It is now possible to choose the type of index for the Encoder and DPR retrievers in order to process the largest corpora while using the GPU.

- Python
Published by raphaelsty over 3 years ago

cherche - 0.0.9

Voting operator dedicated to retrievers and rankers.

- Python
Published by raphaelsty almost 4 years ago

cherche - 0.0.8

Avoid checking similarities in TF-IDF retrievers while filtering documents.

- Python
Published by raphaelsty almost 4 years ago

cherche - 0.0.7

• Significant improvement in the speed of the TF-IDF retriever using sparse CSC matrix.
• The setup.py file loads the readme file as UTF-8.

- Python
Published by raphaelsty almost 4 years ago

cherche - 0.0.6

• Update documentation
• Update retriever Encoder and DPR, path is optionnal
• Add deployment documentation
• Update similarity type
• Avoid round similarity

- Python
Published by raphaelsty almost 4 years ago

cherche - 0.0.5

• Loading and Saving tutorial
• Fuzzy retriever
• Similarities everywhere (retrievers, union, intersection provide similarity scores)
• RAG generation

- Python
Published by raphaelsty about 4 years ago

cherche - 0.0.4

Update of the encoder retriever and the DPR retriever. Documents in the Faiss index will not be duplicated. Query embeddings can now be pre-computed for ranker Encoder and ranker DPR to speed up evaluation without having to compute it again.

- Python
Published by raphaelsty about 4 years ago

cherche - 0.0.3

Adding:

• Translation
• DPR retriever

- Python
Published by raphaelsty about 4 years ago

cherche - 0.0.2

Update of the Cherche dependencies. The previous dependencies were too strict and restrictive as they were limited to a specific version for each package.

- Python
Published by raphaelsty about 4 years ago

cherche - 0.0.1

- Python
Published by raphaelsty about 4 years ago