Only include lint-staged in top-level package.json as outlined in docs:
https://github.com/okonet/lint-staged#how-to-use-lint-staged-in-a-multi-package-monorepo
Moves lint-staged config to top-level package.json since the same
configuration works for all workspaces. lint-staged discovers the closest
configuration to each staged file, so workspace specific configs can be
added in the future if there are conflicts with the global config.
* Remove serpapi dependency given it causes issues with jest, and this is just a regular http call anyway
* Remove it also from docs, why was this here in the first place
* Make CI output easier to understand by splitting up the tasks
* Separate ci into separate tasks
* Try again
* Run build explicitly in test task
* Exclude slow docs build from test tasks
* Try to fix filter on windows
* Use a more reasonable syntaz highlight theme (the default one does not highlight keywords like import and await)
* Small update to Install instructions
* Add yarn/npm install instructions for any place with a npm install command
* Update contributing doc
* Lint
* Add some more cross links and add an index doc for document loaders
* Add guide on which vector store to pick
* Separate install from getting started
* Lint
* Lint
* add base chat models
* starter code for new chains
* Fix tsc issues
* WIP, terrible names (#190)
* add `ChatPrompt` (#197)
* wip add chat prompt template
* add tests
* add export to index
* Refactor LLMChain, add ChatModelChain (#198)
* cr
* add chat model chain
* add load from chat model (#204)
* add load from chat model
* add to chatvectordbchain
* cr
* cr
* fix comment
* use PromptValue
* clear
* fix openai chat model
* llm chain
* fix stuff prompt
* fix build by commenting out old stuff
* format
* consolidate llmresult
* cr
* fix prompts
* rm legacy chains
* export ChatOpenAI
* test and fix chatopenai
* base language model
* refactor schema
* test generate prompt
* revert chains
* format
* fix llm-chain
* add getNumTokens
* add int test for llm chain chat openai
* After merge
* Format
* Add new entrypoints
* Remove console.log
* Remove docs for deprecated model
* Do not clear build artifacts when running tests
* Add docs for chat models
* Add a way of including examples in docs without copy-paste
* hide toc in examples
* Add agent example
* Fix docs build on windows
* Add llm chain example
* Do some more of the serde work
* Add shorthand callPrompt method
* Use .from_ methods to construct prompts
* cr
* cr
* Add shorthand fromTemplate method to chatprompts
---------
Co-authored-by: Nuno Campos <nuno@boringbits.io>
Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
* Use fetch adapter for openai axios
* Remove node-fetch, add instructions for node 16
* Add GH (manual) action for running integration tests (#119)
* Add integration tests GH action (manual trigger for now), reduce cost of integration tests by using smallest possible models
* Convert library to ESM codebase, ESM output (#124)
* Use fetch adapter for openai axios
* Update tsc build to output ESM only
* Update all import paths to have extension per ESM requirements
* Move all source files to src/
* Remove circular dependency
* Fix jest config for ESM
* Remove circular deps not possible with ESM
* Throw error in index getter to dedupe code and have safer usage (#35)
* Fix hnsw for esm
* Fix usage of hnswlib with index passed in
* Fix textsplitter for esm
* Fix openai for esm
* Fix hf for esm
* Fix ESM for cohere
* Fix ESM for serpapi
* Fix esm in srt
* Remove dependency on @vespaiach/axios-fetch-adapter which has an incorrect export
* Fix examples for esm
* Fix entrypoints
* Fix test-exports for esm
* Add fetch flag for node 16 ci job
* Add a more thorough test for packaging
* Fix docs build
---------
Co-authored-by: micahriggan <micahriggan@users.noreply.github.com>
---------
Co-authored-by: micahriggan <micahriggan@users.noreply.github.com>
* Fix one more import
* Fix example
* Also build docs in ci
* Fix sql test in ci
---------
Co-authored-by: micahriggan <micahriggan@users.noreply.github.com>
* Add CI job
* Add ci script
* Run format in examples package
* Format docs
* Fix eslint config in langchain
* Add ci command
* Separate unit and integration tests
* Fix ci command
* Make path relative, adjust node version
* Add os matrix
* Set eol to lf
* Do not cache CI command
* Try to fix loadFromHub on windows
* Update CONTRIBUTING for unit/int tests
* fix: add initial documentation with typedoc
* feat: add docusaurus (#17)
* feat: add docusaurus
* fix: update docs
* Cleanup docs (#19)
* Create Guides category, sort out sidebars
* remove generated files
* Set up ghpages deployment
* Change deployment method
* Temporarily make it run on any push
* Fix api docs broken links, clean up the api docs output a bit
* Revert "Temporarily make it run on any push"
This reverts commit 423c15686c675e9410e311760f7c7934071eb84c.
---------
Co-authored-by: Nuno Campos <nuno@boringbits.io>
* Harrison/add getting started (#20)
* fix: add initial documentation with typedoc
* feat: add docusaurus
* fix: update docs
* Create Guides category, sort out sidebars
* remove generated files
* Set up ghpages deployment
* Change deployment method
* Temporarily make it run on any push
* feat: add docusaurus (#17)
* feat: add docusaurus
* fix: update docs
* Cleanup docs (#19)
* Create Guides category, sort out sidebars
* remove generated files
* Set up ghpages deployment
* Change deployment method
* Temporarily make it run on any push
* Fix api docs broken links, clean up the api docs output a bit
* Revert "Temporarily make it run on any push"
This reverts commit 423c15686c675e9410e311760f7c7934071eb84c.
---------
Co-authored-by: Nuno Campos <nuno@boringbits.io>
* cr
* cr
* cr
* cr
* cr
* fix: remove max-len lint rule
* cr
---------
Co-authored-by: Sean Sullivan <ssullivan61198@gmail.com>
Co-authored-by: Nuno Campos <nuno@boringbits.io>
Co-authored-by: Sean Sullivan <22581534+sullivan-sean@users.noreply.github.com>
* cr
---------
Co-authored-by: Nuno Campos <nuno@boringbits.io>
Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
Nicolas