diff options
| author | Yuren Hao <yurenh2@illinois.edu> | 2026-08-05 19:14:33 -0500 |
|---|---|---|
| committer | Yuren Hao <yurenh2@illinois.edu> | 2026-08-05 19:14:33 -0500 |
| commit | 2de7789ecf05eba74cc52dbd5f4d8298e4084e50 (patch) | |
| tree | 5546f171017966ea7294c4c09a67e4a09778a891 /ep_run/casc_eq_train.py | |
| parent | 0df61c1d36faf3e985394c943853b5bab5adaa8c (diff) | |
PSGD-QUAD入列: vendor psgd.py + Optimizer子类适配器(合成损失closure桥接EP预填梯度), 13路全量回归冒烟PASS
修复过程记录: 首版适配器非Optimizer子类被LambdaLR拒; 重写时splice错序造成muon.py重复段
(旧类后定义胜出), 按结构图手术去重并断言全符号在位。三轴定位: GPU配方候选(追踪慢变统计=
R4幸存哲学+噪声鲁棒卖点), 硬件不合适(d²记忆), 偏置公理最差档(逐方向白化放大低方差偏置)。
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FAPDWQ49M5Ye3NpTndTpn
Diffstat (limited to 'ep_run/casc_eq_train.py')
| -rw-r--r-- | ep_run/casc_eq_train.py | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/ep_run/casc_eq_train.py b/ep_run/casc_eq_train.py index d19504d..db6d71a 100644 --- a/ep_run/casc_eq_train.py +++ b/ep_run/casc_eq_train.py @@ -22,7 +22,7 @@ ap.add_argument('--wandb_run', default='') ap.add_argument('--kmax', type=int, default=8) # adaptive fb rounds cap ap.add_argument('--noguard', action='store_true') # diagnosis: skip only non-finite grads ap.add_argument('--untie', action='store_true') # separate readout matrix (untied from tok) -ap.add_argument('--opt', choices=['adamw', 'muon', 'sgdm', 'lion', 'olion', 'adafactor', 'signline', 'conslion', 'ditherlion', 'cautlion', 'olionns', 'olionk1', 'olionk2', 'olionk3'], default='adamw') +ap.add_argument('--opt', choices=['adamw', 'muon', 'sgdm', 'lion', 'olion', 'adafactor', 'signline', 'conslion', 'ditherlion', 'cautlion', 'olionns', 'olionk1', 'olionk2', 'olionk3', 'psgdquad'], default='adamw') ap.add_argument('--muon_lr', type=float, default=0.02) ap.add_argument('--tok_init', type=float, default=0.0) # >0: init tok/pos with this std (GPT-standard 0.02) ap.add_argument('--compile', action='store_true') # torch.compile each block (free speed where supported) @@ -345,7 +345,7 @@ if args.bf16: if args.untie: with torch.no_grad(): W_out.data = W_out.data.to(torch.bfloat16) print('[bf16] model cast to bfloat16 (E-accum + sigma stay fp32)', flush=True) -if args.opt in ('sgdm', 'lion', 'olion', 'adafactor', 'signline', 'conslion', 'ditherlion', 'cautlion', 'olionns', 'olionk1', 'olionk2', 'olionk3'): +if args.opt in ('sgdm', 'lion', 'olion', 'adafactor', 'signline', 'conslion', 'ditherlion', 'cautlion', 'olionns', 'olionk1', 'olionk2', 'olionk3', 'psgdquad'): from muon import build_alt opt, sched = build_alt(args.opt, blocks, all_params, args.lr, args.warmup, total_steps=(args.steps if args.cosine else 0), lr_min_ratio=args.lr_min_ratio, |