• Joined on 2024年11月12日
domschl pushed to main at domschl/biograph 2026年07月10日 10:48:10 +02:00
08888dcb3e update id and resource uri support
domschl pushed to main at domschl/LatentCalc 2026年07月09日 19:08:51 +02:00
c6c804c241 Document latent Gaussian noise injection in README.md
9ff22f3dd8 Implement latent Gaussian noise injection during training for DEQ stabilization
Compare 2 commits »
domschl pushed to main at domschl/LatentCalc 2026年07月09日 18:52:13 +02:00
d13682243e Slower epsilon decay
domschl pushed to main at domschl/LatentCalc 2026年07月09日 18:34:47 +02:00
28d770c8a3 Document epsilon curriculum in README.md
domschl pushed to main at domschl/LatentCalc 2026年07月09日 18:34:30 +02:00
7493946b93 Introduce dynamic epsilon curriculum progression starting at 0.1
domschl pushed to main at domschl/LatentCalc 2026年07月09日 18:12:51 +02:00
8435490c6e Document compilation flags in README.md
012d26834c Default --compile to True on Linux, and add overriding --no-compile flag
61306c812d Document Anderson Acceleration solver in README.md
6cc317d1e4 Implement MPS-compatible Tikhonov-regularized Anderson Acceleration solver for DEQ models
Compare 4 commits »
domschl pushed to main at domschl/LatentCalc 2026年07月09日 17:14:49 +02:00
b1ec23b1a7 Document the project in README.md
domschl pushed to main at domschl/LatentCalc 2026年07月09日 17:09:25 +02:00
19e019b15b Add selectable core-type option, default to transformer, and enable TF32 matmul precision
domschl pushed to main at domschl/LatentCalc 2026年07月09日 16:54:37 +02:00
be6d595626 Integrate pure PyTorch MambaBlock as core and add configurable Taylor-order backpropagation
domschl pushed to main at domschl/LatentCalc 2026年07月09日 15:54:54 +02:00
a99bb1b728 Add orthogonal initialization, core spectral normalization, weight decay, and split core learning rates
domschl pushed to main at domschl/LatentCalc 2026年07月09日 15:21:16 +02:00
c84c31152d Add --n-fixed-core configurable depth parameter for the recurrent reasoning core
domschl pushed to main at domschl/LatentCalc 2026年07月09日 15:04:46 +02:00
e9e78d07a2 Add --epsilon CLI parameter and log FP-Diff in train.py
eafd79f05a Dynamically label print metric as FP-Iterations in DEQ mode
Compare 2 commits »
domschl pushed to main at domschl/LatentCalc 2026年07月09日 14:32:05 +02:00
088a76919f Update .python-version to 3.14
domschl pushed to main at domschl/LatentCalc 2026年07月09日 14:30:57 +02:00
9191a353f7 Add numpy dependency
domschl pushed to main at domschl/LatentCalc 2026年07月09日 14:29:42 +02:00
12d5ecb096 Support logging actual steps count in DEQ model mode
domschl pushed to main at domschl/LatentCalc 2026年07月09日 14:22:51 +02:00
3844d6fd7e DEQ tests
domschl pushed to main at domschl/LatentCalc 2026年07月09日 13:25:21 +02:00
21ac79f0e2 Initialize project: Sandwich Universal Transformer with ACT in PyTorch
domschl created branch main in domschl/LatentCalc 2026年07月09日 13:25:21 +02:00
domschl created repository domschl/LatentCalc 2026年07月09日 13:13:18 +02:00
domschl pushed to main at domschl/biograph 2026年07月07日 09:04:47 +02:00
7a8019d4c1 Client: display calibre books added