Skip to main content

LLM that predicts text diffs.

Project description

DiffLLM

Extending the next token prediction setting into "diff prediction".

Setup

Create virtual environment:

conda create -n diff-llm python=3.9
conda activate diff-llm
pip install -r requirements.txt

Export secrets:

export $(grep -v '^#' secrets.txt | xargs)

Export env vars:

export PYTHONPATH=.

Usage

Create dataset

python src/create_dataset.py \
    --output-dir ./dataset \
    --page-names '["Deep learning", "Ancient Greece", "Ted Chiang"]' \
    --n-revisions 10

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

diff-llm-0.0.0a0.tar.gz (6.3 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

diff_llm-0.0.0a0-py3-none-any.whl (7.6 kB view details)

Uploaded Python 3

File details

Details for the file diff-llm-0.0.0a0.tar.gz.

File metadata

  • Download URL: diff-llm-0.0.0a0.tar.gz
  • Upload date:
  • Size: 6.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.9.17

File hashes

Hashes for diff-llm-0.0.0a0.tar.gz
Algorithm Hash digest
SHA256 21891dc39bc0d390ecc9aa2c50d4d4fc9b32320f5e22ece35d8c5d1400578089
MD5 ac652627646afaa0cb0cce8bfb5d4947
BLAKE2b-256 bb8504dcd9a8525790886c335a2252c848220be291673c0cbf64359d7b9f1e55

See more details on using hashes here.

File details

Details for the file diff_llm-0.0.0a0-py3-none-any.whl.

File metadata

  • Download URL: diff_llm-0.0.0a0-py3-none-any.whl
  • Upload date:
  • Size: 7.6 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.9.17

File hashes

Hashes for diff_llm-0.0.0a0-py3-none-any.whl
Algorithm Hash digest
SHA256 fbe8a8d60173b2a7bf53973c2d9d82c54dda5a91a41ff39b62a9233da719ac20
MD5 a0d67b5466384bce384ce39bb44e5733
BLAKE2b-256 683d3249a4b82dac47cf314164643361d9dcf9188bebfbb81452904db55773e0

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page