7 Commits

Author	SHA1	Message	Date
SikongJueluo	4ea567adba	feat(compressors): add JSONL training metrics logging with CLI controls ... - Add write_training_metrics() in new compressors/training_metrics.py for appending epoch/step/lr/component rows as JSON Lines - Wire --metrics-path and --log-every CLI options into train.py, passing them to the training loop so metrics rows are written every N steps - Accept absolute metrics paths or paths relative to output directory - Add quantization component to loss log alongside existing distill/contrastive - Replace inline torch.device() with get_device() utility - Add test_hash_training_metrics.py covering multi-row JSONL append Infrastructure: - Pin torch 2.7.1 + CUDA 12.8 index for Linux/Windows in pyproject.toml - Add .justfile rsync upload recipe with .stignore exclusion - Exclude **/__marimo__ from rsync in .stignore Dependencies updated: numpy 2.4.5, pandas 3.0.3, black 26.5.0, click 8.4.0, contourpy, etc.	2026-05-17 14:57:10 +08:00
SikongJueluo	e832f9d656	refactor(ui): replace tqdm with rich for enhanced console output	2026-03-06 17:07:59 +08:00
SikongJueluo	f61857feba	fix(compressors): fix the wrong usage of loss function in training pipeline	2026-02-28 17:51:14 +08:00
SikongJueluo	1926cb53e2	feat(compressors): replace float/int compressors with hash-based compression for CAM	2026-02-24 22:54:41 +08:00
SikongJueluo	7f6732edeb	feat(benchmarks): add evaluation framework for DINO-based compressors	2026-02-10 11:19:17 +08:00
SikongJueluo	76a572ee12	feat(train): add checkpointing and training interruption handling	2026-02-08 17:28:19 +08:00
SikongJueluo	8f417b674c	feat(compressors): add neural compression modules and training pipeline	2026-02-08 16:41:34 +08:00