1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
|
"""Ship-gate for holo_a_track_avg (trend-aware stop + plateau averaging, the semi-convergence fix):
cos(EP,BPTT) at t2sel=160 on s2000 — the regime where argmin-t_best got fooled (t2_probe: batch2
degraded 0.956->0.894 at 160). PASS = avg recovers the t2sel-80 level (~0.93+) at 160 without
hurting the healthy batches."""
import torch
import lt_ep_train as L
from diag_cos import cos_ep_bptt
import holo_ep
torch.manual_seed(0)
blk = L.EQBlock(512, 16, 256, 256, c=1.0, attn_mode='thick'); blk.qknorm = True
ck = torch.load('runs/redx_traj/s2000.pt', map_location=L.dev)
with torch.no_grad():
for p, w in zip(blk.allp, ck['allp']):
p.copy_(w.to(L.dev))
blk.track = True
blk.holofast = True
torch.manual_seed(11)
batches = [L.get_batch('train', 24, 256) for _ in range(3)]
orig_fast = holo_ep.holo_a_track_fast
for name, fn in (('argmin(fast)', holo_ep.holo_a_track_fast), ('trend+avg', holo_ep.holo_a_track_avg)):
holo_ep.holo_a_track_fast = fn # ep_step routes track via holofast -> this symbol
import lt_ep_train
lt_ep_train.ep_step.__globals__ # (route happens at import inside ep_step)
cs = []
for idx, y in batches:
c, _ = cos_ep_bptt(blk, idx, y, 150, 20, 0.1, 0.02, holo=2, hr=0.02, t2sel=160, bsub=4)
cs.append(c)
print(f"{name:>13}: cos={' '.join(f'{c:.4f}' for c in cs)} mean={sum(cs)/len(cs):.4f}", flush=True)
holo_ep.holo_a_track_fast = orig_fast
print("NEWSTOP_GATE_DONE", flush=True)
|
