Justin Bollinger f0bba73225 fix: auto-detect training device instead of defaulting to CUDA ...

The PassGPT training device menu now uses _detect_device() to default
to the best available device (CUDA > MPS > CPU) rather than always
defaulting to CUDA, which fails on systems without NVIDIA GPUs.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-02-18 18:47:41 -05:00

..

__init__.py

fix: clean version string to show 2.0+g<hash> format

2026-02-13 22:11:31 -05:00

__main__.py

Restore hate_crack package and add E2E install tests

2026-01-31 22:50:51 -05:00

api.py

fix: resolve Hashview wordlist downloads to configured directory

2026-02-17 11:05:23 -05:00

attacks.py

fix: auto-detect training device instead of defaulting to CUDA

2026-02-18 18:47:41 -05:00

cli.py

fix: --debug flag now outputs HTTP requests to screen

2026-02-18 15:17:36 -05:00

config.json.example

feat: auto-bump patch version on PR merge to main

2026-02-18 14:58:47 -05:00

formatting.py

updated formatting to take account terminal width

2026-02-03 17:07:26 -05:00

main.py

fix: set HF_HUB_DISABLE_TELEMETRY before HuggingFace imports in main.py

2026-02-18 15:34:11 -05:00

passgpt_generate.py

fix: pass --debug flag to PassGPT subprocesses and create GitHub Releases on version bump

2026-02-18 15:26:53 -05:00

passgpt_train.py

fix: pass --debug flag to PassGPT subprocesses and create GitHub Releases on version bump

2026-02-18 15:26:53 -05:00