v0.2.0 #48
Draft
v0.2.0 #48
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This directory will be uploaded in the future during testing
…ll object. Currently, it's only implemented for topk parsing
* added support to new chatmodels * Ignored paper directory to avoid commiting datasets by mistake This directory will be uploaded in the future during testing * Fixed imports to be compatible with the new version of langchain * fixed typo in vectostores * Implemented LLM model clases and made them independent from the AskTell object. Currently, it's only implemented for topk parsing
* added support to new chatmodels * Ignored paper directory to avoid commiting datasets by mistake This directory will be uploaded in the future during testing * Fixed imports to be compatible with the new version of langchain * fixed typo in vectostores * Implemented LLM model clases and made them independent from the AskTell object. Currently, it's only implemented for topk parsing * Exposed to control the diversity in MMR selection * updated inverse design prompt and test notebooks
Inverse lambda
* updated gitignore file * Exposed system messages to the user The prompt can be specified using the `inv_system_message` and the `system_message` arguments in the `ask`, `predict`, and `inv_predict` methods."
TODO: solve the problem with anyscale only accepting requests with n=1
* Started implementation of LLM-specific tests * Updated tests * Updated AskTell tests * fixed logprobs type in OpenAILLM * Updated chat models to use logprobs * removed unneeded imports * Updated tests TODO: implements tests for KNN, KRR, and finetune
* Updated regression experiments notebook * Started implementation of LLM-specific tests * Updated tests * fixed conflicts after merging the update_tests branch * Cleaned exps notebooks and ocm generator notebook * Created a base class for AskTellFewshot Children classes should implement specific hidden methods * Created a base class for AskTellFewshot Children classes should implement specific hidden methods * Updated AskTellGPR to inherit from AskTellFewShot * Updated AskTellGPR to inherit from AskTellFewShot * Updated AskTellNearestNeighbor to inherit from AskTellFewShot * Updated tests and removed old tests_bolift.py
* Changed default top_p to 0.5 * processed the string being trained with the class formatters * updated regression scripts and added results * Updated runs for OCM
* Changed default top_p to 0.5 * processed the string being trained with the class formatters * updated regression scripts and added results * Updated runs for OCM * Fixed the pydantic error of BO-LIFT tool * Changed hor logprobs is passed to langchain in new version Depending on the version of langchain/openai, the logprobs argument needs to passed as kwargs or explicitly. I changed it to be compatible with langchain 0.3.1, langchain-openai 0.2.1, and openai 1.51. TODO: constrain these versions in the setup.py * Updated experiments notebooks * updated gitignore and removed unneeded committed files * updated results data
* Changed default top_p to 0.5 * processed the string being trained with the class formatters * updated regression scripts and added results * Updated runs for OCM * Fixed the pydantic error of BO-LIFT tool * Changed hor logprobs is passed to langchain in new version Depending on the version of langchain/openai, the logprobs argument needs to passed as kwargs or explicitly. I changed it to be compatible with langchain 0.3.1, langchain-openai 0.2.1, and openai 1.51. * Updated experiments notebooks * started claude implementation TODO: Anthropic doesnt have the option to generate multiple completions such as openAI. I need to implement it myself * Finished implementation of claude multi-request * Added langchain-anthropic as a requirment * Added claude models in the functions to load tests (get_dataset and get_asktell)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.