diff options
| author | YurenHao0426 <blackhao0426@gmail.com> | 2026-01-13 23:50:59 -0600 |
|---|---|---|
| committer | YurenHao0426 <blackhao0426@gmail.com> | 2026-01-13 23:50:59 -0600 |
| commit | 00cf667cee7ffacb144d5805fc7e0ef443f3583a (patch) | |
| tree | 77d20a3adaecf96bf3aff0612bdd3b5fa1a7dc7e /runs/slurm_logs/14632871_test_opt.out | |
| parent | c53c04aa1d6ff75cb478a9498c370baa929c74b6 (diff) | |
| parent | cd99d6b874d9d09b3bb87b8485cc787885af71f1 (diff) | |
Merge master into main
Diffstat (limited to 'runs/slurm_logs/14632871_test_opt.out')
| -rw-r--r-- | runs/slurm_logs/14632871_test_opt.out | 29 |
1 files changed, 29 insertions, 0 deletions
diff --git a/runs/slurm_logs/14632871_test_opt.out b/runs/slurm_logs/14632871_test_opt.out new file mode 100644 index 0000000..5478903 --- /dev/null +++ b/runs/slurm_logs/14632871_test_opt.out @@ -0,0 +1,29 @@ +Testing optimized SpikingVGG forward... +Testing on device: cuda +Using: Global delta + Global renorm (Option 1 - textbook LE) +Model depth: 6 conv layers +Parameters: 1,187,274 + +[Test 1] Forward without Lyapunov... + Logits shape: torch.Size([8, 10]) ✓ + +[Test 2] Forward with Lyapunov... + Logits shape: torch.Size([8, 10]) ✓ + Lyapunov exponent: 3.3281 ✓ + +[Test 3] Backward pass... + Loss: 5.3691 ✓ + Gradient norm: 25.8283 ✓ + No NaN gradients ✓ + +[Test 4] Multiple training steps... + Step 1: loss=7.0680, λ=3.0538 + Step 2: loss=7.3570, λ=3.2788 + Step 3: loss=8.4584, λ=2.8295 + Step 4: loss=5.7036, λ=3.3123 + Step 5: loss=7.4161, λ=3.4701 + +================================================== +ALL TESTS PASSED! +================================================== +Done! |