All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog.
- Correct invalid dependency of promptlayer slipped in the build
- Scroll to the first relevant context passage, if the most relevant context passage is at the end, it will scroll to the end of the document
- Added Mistral NEMO as default model
- Rearranged the interface to get more space
- Updated libraries to the latest versions
- Fixed the chat messages sequence that were buggy
- Updated the PDF viewer to the latest version
- Add selection of embedding functions
- Add selection of text from the pdf viewer (provided by https://github.com/lfoppiano/streamlit-pdf-viewer)
- Added an experimental feature for calculating the coefficient that relate the question and the embedding database
- Added the data availability statement in the searchable text
- Removed obsolete and non-working models zephyr and mistral v0.1
- The underlying library was refactored to make it easier to maintain
- Removed the native PDF viewer
- Updated langchain and streamlit to the latest versions
- Removed conversational memory which was causing more problems than bringing benefits
- Rearranged the interface to get more space
- Updated and removed models that were not working
- Fixed problems with langchain and other libraries
- Add gpt4 and gpt4-turbo
- improved UI: replace combo boxes with dropdown box
- Fixed dependencies when installing as library
- Add experimental PDF rendering in the page
- Fix GrobidProcessors API implementation
- Remove memory when using Zephyr-7b-beta, that easily hallucinate
- Include biblio in embeddings by @lfoppiano in #21
- Fix conversational memory by @lfoppiano in #20
- add zephyr-7b by @lfoppiano in #15
- add conversational memory in #18
- fix env variables by @lfoppiano in #9
- Selection of chunk size on which embeddings are created upon
- Mistral model to be used freely via the Huggingface free API
- Improved documentation, adding privacy statement
- Moved settings on the sidebar
- Disable NER extraction by default, and allow user to activate it
- Read API KEY from the environment variables and if present, avoid asking the user
- Avoid changing model after update
- ChromaDb accumulating information even when new papers were uploaded
- docker build
- Github action build
- dependencies of langchain and chromadb
- pypi package
- docker package release
- Kick off application
- Support for GPT-3.5
- Support for Mistral + SentenceTransformer
- Streamlit application
- Docker image
- pypi package