timmy-config

Author	SHA1	Message	Date
Claude (Opus 4.6)	099948b3d1	fix: use PYTHON variable in training Makefile Some checks failed Architecture Lint / Linter Tests (pull_request) Successful in 8s Details PR Checklist / pr-checklist (pull_request) Failing after 2m12s Details Smoke Test / smoke (pull_request) Failing after 13s Details Validate Config / YAML Lint (pull_request) Failing after 9s Details Validate Config / JSON Validate (pull_request) Successful in 10s Details Validate Config / Python Syntax & Import Check (pull_request) Failing after 16s Details Validate Config / Shell Script Lint (pull_request) Failing after 15s Details Validate Config / Cron Syntax Check (pull_request) Successful in 7s Details Validate Config / Deploy Script Dry Run (pull_request) Successful in 7s Details Validate Config / Playbook Schema Validation (pull_request) Successful in 13s Details Architecture Lint / Lint Repository (pull_request) Has been cancelled Details Validate Config / Python Test Suite (pull_request) Has been cancelled Details On macOS where only python3 is installed (no python shim), bare `python` calls fail with 'No such file or directory'. Adds `PYTHON ?= python3` variable. Replaces all bare `python` calls with `$(PYTHON)` across: train-local, vibes, adversary-value-violations, ingest, curated, convert. Override: make vibes PYTHON=python Closes #660 Refs Timmy_Foundation/the-nexus#1471	2026-04-17 06:39:05 +00:00
Claude (Opus 4.6)	ef58883a26	fix: use PYTHON variable in training Makefile for portability (closes #660 ) Added PYTHON ?= python3 variable and replaced all bare python calls. Fixes macOS where only python3 is installed. Refs #660	2026-04-17 02:37:47 -04:00
Merge Bot	a27061b38b	Merge PR #653 : training/Makefile (changed)	2026-04-16 05:12:07 +00:00
Merge Bot	fb5f2908e5	Merge PR #760 : training/Makefile (changed)	2026-04-16 05:07:27 +00:00
perplexity	6507cffc15	feat: migrate autolora pipeline into training/ Per direction shift (the-nexus#542). Replaces the autolora repo (1,500 lines of custom pipeline code) with config files for existing tools: - axolotl.yaml: replaces train_modal.py (239 lines) - mlx-lora.yaml: replaces MLX training scripts - eval-tasks.yaml: replaces run_eval.py (300 lines) - Makefile: replaces run_vibes.py, compare.py, convert_to_mlx.py Data migrated as-is: - curated_dataset.jsonl (26 gold-standard conversations) - preference_pairs.jsonl (DPO pairs) - prompts_vibes.yaml, prompts_nexus_vibes.yaml - v0-baseline eval results (historical record) Thin glue kept: - build_curated.py (data authoring, not infrastructure) - ingest_trajectories.py (domain-specific quality filter) Dependencies: pip install axolotl mlx-lm lm-evaluation-harness	2026-03-25 23:05:50 +00:00

5 Commits