alan-turing-institute/reginald

Issues

• `poetry.lock` Continuous Integration with Dependabot

  #201 opened 5 months ago by griff-rees
  0

• Implement clear history for the API model

  #199 opened 5 months ago by rwood-97
  1

• Try out Mixtral-8x7B model

  #149 opened 5 months ago by rchan26
  1

• Update pulumi script(s)

  #113 opened 5 months ago by rchan26
  3

• Refactor library to avoid too many imports

  #193 opened 5 months ago by rchan26
  0

• Create CLI for Reginald

  #159 opened 5 months ago by rchan26
  1

• Add streaming option

  #188 opened 5 months ago by rchan26
  0

• Use Github actions to launch pulumi to update models

  #140 opened 5 months ago by rchan26
  0

• Reginald production workflow

  #127 opened a year ago by rchan26
  0

• Add Ollama model

  #179 opened 5 months ago by rchan26
  1

• Add try/except with max number of retries for building index

  #157 opened 6 months ago by rchan26
  0

• Add loading of Huggingface token

  #172 opened 8 months ago by rchan26
  0

• Update Reginald with new llama-index

  #162 opened 8 months ago by rwood-97
  0

• Update collection of turing.ac.uk data

  #153 opened 10 months ago by rwood-97
  0

• Investigate using Github secrets for storing API keys

  #130 opened a year ago by rchan26
  0

• Containerise data index creation

  #143 opened a year ago by rchan26
  0

• Add ping/pong for finding if the VM is on (i.e. if Reginald is OOO)

  #145 opened a year ago by rwood-97
  0

• Add better README for pulumi stuff

  #133 opened a year ago by rwood-97
  0

• Better chat history management

  #101 opened a year ago by rchan26
  3

• Add GIF of asking Reginald a question on Slack to README

  #129 opened a year ago by rchan26
  0

• Re-explore use of agent-based chat engine

  #126 opened a year ago by rwood-97
  0

• Updating data indexes

  #124 opened a year ago by rchan26
  1

• Add `mypy` to pre-commit

  #108 opened a year ago by rchan26
  0

• Add try-except to loading in the data index

  #123 opened a year ago by rchan26
  0

• Make LlamaIndexHF default model microsoft/phi-1_5

  #80 opened a year ago by rchan26
  1

• Update README

  #117 opened a year ago by rchan26
  0

• Make emoji an CLI argument for initialising the full Reginald model

  #120 opened a year ago by rchan26
  0

• Reorganise repo

  #106 opened a year ago by rchan26
  1

• Merge direct_message and channel_mention methods into response

  #107 opened a year ago by rchan26
  0

• Ensure slackbot doesn't crash when multiple messages are sent at the same time by different users

  #94 opened a year ago by rwood-97
  2

• Docker script for running slack bot

  #112 opened a year ago by rchan26
  0

• Clean up pyproject.toml

  #104 opened a year ago by rchan26
  0

• Fix docker for running "full" model

  #111 opened a year ago by rchan26
  0

• Set up API between container and LLM

  #95 opened a year ago by rchan26
  1

• Add CLI shortcuts

  #105 opened a year ago by rchan26
  0

• Only add direct_message and channel_mention events to queue

  #100 opened a year ago by rchan26
  1

• Send different emoji depending on what place it is in the queue

  #99 opened a year ago by rchan26
  1

• Add ability to clear chat history in slack bot

  #88 opened a year ago by rwood-97
  2

• Figure out how to ensure messages are emoji'd when in queue

  #90 opened a year ago by rwood-97
  1

• Update data preparation in LlamaIndex

  #82 opened a year ago by rchan26
  1

• Figure out how to have multiple chat instances for different users

  #86 opened a year ago by rchan26
  0

• Decide which arguments should be set using defaults

  #87 opened a year ago by rwood-97
  2

• Implement queuing system for query slack bot

  #83 opened a year ago by rwood-97
  6

• Add Llama-2 model to existing available query engines

  #78 opened a year ago by rchan26
  0

• Fine-tune embedding model (by using adapters)

  #84 opened a year ago by rchan26
  0

• Create llama-index reader for Hut23 issues

  #72 opened a year ago by rchan26
  2

• Replace LLMPredictor with LLM in ServiceContext creation

  #74 opened a year ago by rchan26
  0

• Replace GPTVectorStoreIndex with VectorStoreIndex

  #75 opened a year ago by rchan26
  0

• Replace CustomLLM with llama-index's HuggingfaceLLM

  #73 opened a year ago by rchan26
  0

• Use llama-hub reader to read data directly from GitHub wikis

  #77 opened a year ago by rwood-97
  1