amaiya/onprem

Issues

• Add preproc_fn to Extractor.apply

  #74 opened 3 months ago by amaiya
  0

• PDF OCR support

  #75 opened 3 months ago by amaiya
  0

• It may be possible to reduce docker image size significantly

  #73 opened 3 months ago by dwvisser
  0

• Remove BOS token from default prompt

  #67 opened 5 months ago by amaiya
  0

• Segment needs to accept arguments in extractor pipeline

  #70 opened 5 months ago by mzientek
  1

• Add clean function to Extractor.apply

  #69 opened 5 months ago by amaiya
  0

• Remove call to persist

  #68 opened 5 months ago by amaiya
  0

• Few-Shot classification pipeline

  #66 opened 5 months ago by amaiya
  0

• add check for partially download files

  #49 opened 5 months ago by amaiya
  0

• change default model to Mistral

  #65 opened 5 months ago by amaiya
  0

• information extraction pipeline

  #64 opened 5 months ago by amaiya
  0

• experimental support for Azure OpenAI

  #63 opened 5 months ago by amaiya
  0

• allow installation of onprem without llama-cpp-python for easier use with LLMs served through REST APIs

  #62 opened 5 months ago by amaiya
  0

• Use OnPrem.LLM with OpenAI-compatible REST APIs

  #61 opened 5 months ago by amaiya
  0

• Issue warning instead of halting if encountering an error when loading files during ingest

  #60 opened 5 months ago by amaiya
  0

• Add `Ingester.get_ingested_files` to show files ingested in vector databae

  #59 opened 5 months ago by amaiya
  0

• Add ignore_fn argument to LLM.ingest

  #58 opened 5 months ago by amaiya
  0

• Curious

  #57 opened 9 months ago by agentsimon
  2

• ssl issue to get embedding model files for local usage of onprem

  #56 opened 9 months ago by ringo70
  1

• offload_kqv not properly set

  #50 opened 10 months ago by amaiya
  1

• support OpenAI models

  #55 opened 10 months ago by amaiya
  0

• Accept extra **kwargs in prompt and pass to model

  #54 opened 10 months ago by amaiya
  0

• Add `stop` parameter to `LLM.prompt` method

  #53 opened 10 months ago by amaiya
  0

• Use Zephyr-7B as default model in Web app

  #52 opened 10 months ago by amaiya
  0

• Add prompt_template argument to LLM

  #51 opened 10 months ago by amaiya
  1

• add pipeline module

  #35 opened a year ago by amaiya
  0

• add guardrails module

  #34 opened a year ago by amaiya
  0

• CPU limits

  #48 opened a year ago by lystrata
  1

• add python-pptx as dependency

  #44 opened a year ago by amaiya
  0

• have ingest skip ~$ files created by Windows

  #45 opened a year ago by amaiya
  0

• add prompt template from YAML to ask in Web app

  #47 opened a year ago by amaiya
  0

• add progress bar for ingest

  #46 opened a year ago by amaiya
  0

• "No module named 'docx' error

  #43 opened a year ago by amaiya
  0

• Support for LlamaIndex

  #39 opened a year ago by nawagner
  2

• Core dumped / segmentation fault

  #41 opened a year ago by lysa324
  2

• Support for Mistral 7B Model

  #40 opened a year ago by rabilrbl
  2

• Number of tokens

  #38 opened a year ago by lysa324
  3

• ValidationError: 1 validation error for LlamaCpp

  #37 opened a year ago by Jacob-Langley
  3

• support for custom metadata in vectorstore

  #36 opened a year ago by amaiya
  0

• remove pin for `llama-cpp-python` so latest is used

  #33 opened a year ago by amaiya
  0

• include `prompt_template` in YAML for Web app

  #32 opened a year ago by amaiya
  0

• Change `LLM.ask` to return dictionary with keys: `answer`, `source_documents`, and `question`

  #31 opened a year ago by amaiya
  0

• load llm in constructor

  #30 opened a year ago by amaiya
  0

• round scores in web app

  #29 opened a year ago by amaiya
  0

• include hyperlinks to sources

  #28 opened a year ago by amaiya
  0

• add support for `score_threshold` in `LLM.ask` and `LLM.chat`

  #26 opened a year ago by amaiya
  0

• Talk to Your Documents with Retrieval-Augmented Generation not working

  #25 opened a year ago by Velcin
  3

• use `CallbackManager`

  #24 opened a year ago by amaiya
  0

• return `source_documents` used to generate answer in `LLM.chat`

  #23 opened a year ago by amaiya
  0

• support for custom callbacks

  #21 opened a year ago by amaiya
  0