Hugging Face's logo

Spaces:
lukestanley
/
ChillTranslator
Sleeping

App Files Community
1
ChillTranslator / utils.py

Commit History

  1. Add Anthropic Opus support

    lukestanley
    64c61f3

  1. Fix: when using http worker, only download if inference is on localhost

    lukestanley
    52fc709

  1. Add retry logic upon schema fail for Mistral API calls

    lukestanley
    1e622b4

  1. Fix import

    lukestanley
    cf4506f

  1. Add log_to_jsonl function to data.py and remove duplicate function from utils.py

    lukestanley
    44bab49

  2. Add JSONL disk file logging functionality to app.py and utils.py

    lukestanley
    abbebf8

  3. Tidying: reordering code a bit

    lukestanley
    124003a

  4. Add Mistral API support due to my RunPod serverless system reliability issues

    lukestanley
    8093276

  5. Add TODO for Runpod timeout handling

    lukestanley
    3c6c618

  6. Assert RunPod env vars are setup before trying to use them

    lukestanley
    00af17e

  1. Revert expected serverless output metadata stripper

    Luke Stanley
    c013599

  2. Avoid unneeded imports, make serverless output more sensible, removing some debugging and comments

    Luke Stanley
    469f650

  3. Fix RUNPOD_ENDPOINT_ID environment variable

    Luke Stanley
    ce5ad5f

  4. Introduces worker mode env var

    Luke Stanley
    56e785c

  1. Ensure N_GPU_LAYERS is int

    Luke Stanley
    9475016

  1. Update default GPU layer, temperature values

    lukestanley
    e327a9e

  2. Add env vars to set GPU layer count and context size, make verbose

    lukestanley
    e01e28e

  1. Add n_gpu_layers parameter to Llama initialization

    lukestanley
    88e6118

  2. Fix: Move n_ctx parameter to model setup!

    lukestanley
    358cd20

  3. Fix check for LLM_MODEL_PATH to avoid load error

    lukestanley
    ff938c3

  4. Auto-downloads model if env var is not set

    Luke Stanley
    74d6e52

  5. Make llm_stream_sans_network actually stream to stdout

    Luke Stanley
    a0f49a0

  6. Default to in-memory LLM interface

    Luke Stanley
    ddb0d91

  1. Add comment about adaptability

    lukestanley
    6c32632

  2. Print new line after LLM output end and some linting

    lukestanley
    3ebb6e1

  1. Makes URL more obvious, update comments, lowers temp

    lukestanley
    a96b492

  2. Move prompt strings and types to own file, reorder code a bit

    lukestanley
    139217d

  3. Linting with Black

    lukestanley
    e4b918c

  4. Removes unused code

    lukestanley
    811e485

  5. Removed unused lines

    lukestanley
    f84c1a6

  6. Slight refactor tidying

    lukestanley
    327982a