Morph MORPH
SIGN IN SIGN UP
dhruvhead / ao UNCLAIMED
0 0 0 Python

COMMITS

/ test/float8/test_compile.py
docs
August 13, 2025
A
Drop support for PyTorch 2.5 and before (#2720)
andrewor14 committed
3bad6a2
July 12, 2025
J
Rename torchao.float8.Float8Tensor to Float8TrainingTensor (#2535)
Jerry Zhang committed
378e179
June 24, 2025
V
rename `torchao.testing.float8` to `torchao.testing.training` (#2415)
Vasiliy Kuznetsov committed
1a701e6
April 22, 2025
J
Update ruff version to 0.11.6 (#2103)
Jerry Zhang committed
5b775fe
February 22, 2025
V
delete delayed scaling from torchao.float8 (#1753)
Vasiliy Kuznetsov committed
d370196
February 19, 2025
V
float8 training: make the "config from recipe" API polished (#1731)
Vasiliy Kuznetsov committed
c6c388b
V
float8 training: clean up recipe names (#1730)
Vasiliy Kuznetsov committed
7fc8ad4
February 10, 2025
D
Support power of 2 scaling factors in float8 training and use e4m3 everywhere (#1670)
Daniel Vega-Myhre committed
32a51ec
January 16, 2025
Y
Add a register_replacement to fix float8 delayed scaling kernel fusion issues in torchao/float8
y-sq committed
74a15f1
December 19, 2024
V
refactor: move delayed and static scaling to StatefulFloat8Linear (#1439)
Vasiliy Kuznetsov committed
7618c26
December 4, 2024
L
[float8] Allow specifying arbitrary dtype for each tensor (#1378)
Luca Wehrstedt committed
4f8021f
December 1, 2024
A
Update hardware check conditions (#1356)
Apurva Jain committed
22bec74
November 18, 2024
A
Lint fixes for test/float8 (#1303)
Apurva Jain committed
6234116
October 23, 2024
J
[reland][ROCm] use dataclass for fnuz type setting (#1150)
Jeff Daily committed
75d0693
M
Revert "[ROCm] use dataclass for fnuz type setting" (#1148)
Mark Saroufim committed
d252612
J
[ROCm] use dataclass for fnuz type setting (#1142)
Jeff Daily committed
eb1fb3a
October 7, 2024
V
float8 training axiswise scaling support with per-gemm-argument configuration (#940)
Vasiliy Kuznetsov committed
dec0313
V
add axiswise scaling to Float8Linear (#920)
Vasiliy Kuznetsov committed
e76db70
October 1, 2024
W
[float8] improve eager numerics for dynamic scales and gets on par with torch.compile (#904)
Wei (Will) Feng committed
71be315
September 24, 2024
V
clean up device checks in float8 unit test files (#923)
Vasiliy Kuznetsov committed
26e790d
August 28, 2024
V
float8 training: add static scaling (#760)
Vasiliy Kuznetsov committed
983f565
August 15, 2024
M
Add PyTorch 2.4 tests in CI (#654)
Mark Saroufim committed
ffa88a4
August 14, 2024
M
retry version guard fix (#679)
Mark Saroufim committed
1acd710
August 7, 2024
A
Enable float8 CI on sm89 (#587)
Apurva Jain committed
1cfe69e
July 30, 2024
V
move float8_experimental to torchao/float8
Vasiliy Kuznetsov committed
e1aee63