Casey Primozic 52b7105f87 Dedup params in Optimizer (#1047) ...

* Dedup params in optimizer

 * Passing the same tensor multiple times in the set of learnable params passed to optimizers can result in models completely failing to learn, but no errors are produced.  This dedups tensors to avoid the problem.

* Fix types

* Use new variable to satisfy linter

* Use `helpers.dedup` instead of `set()` to dedup params

* Add test for duped params in optimizers

2023-06-26 00:49:23 -07:00

..

external

touchups

2023-06-25 15:24:26 -07:00

extra

Windows & Ubuntu CLANG CI support (#1011)

2023-06-19 09:33:24 -07:00

models

Windows & Ubuntu CLANG CI support (#1011)

2023-06-19 09:33:24 -07:00

unit

test speed llama (#1046)

2023-06-25 15:22:56 -07:00

__init__.py

All devices are equal! (#196)

2020-12-15 23:44:08 -08:00

Dockerfile

Docker fix (#1039)

2023-06-25 10:38:58 -07:00

test_assign.py

Refactor LoadOps (#910)

2023-06-03 09:40:43 -07:00

test_conv_shapetracker.py

speed like you wouldn't believe (#626)

2023-03-02 07:49:19 -08:00

test_conv.py

Add conv binops_no_rerun test assertions (#665)

2023-03-10 19:09:48 -08:00

test_custom_function.py

Disktensors! (#819)

2023-05-28 15:40:37 -07:00

test_dtype.py

add cuda on cpu tests (#1020)

2023-06-22 14:15:50 -07:00

test_gc.py

fix up eye and fix gc test

2023-02-27 06:53:18 -08:00

test_jit.py

assertions for jit

2023-05-05 21:56:32 -07:00

test_net_speed.py

test speed llama (#1046)

2023-06-25 15:22:56 -07:00

test_nn.py

Windows & Ubuntu CLANG CI support (#1011)

2023-06-19 09:33:24 -07:00

test_ops.py

MaskRCNN Inference (#884)

2023-06-25 15:37:51 -07:00

test_optim.py

Dedup params in Optimizer (#1047)

2023-06-26 00:49:23 -07:00

test_randomness.py

Refactor LoadOps (#910)

2023-06-03 09:40:43 -07:00

test_specific_conv.py

disable that test on LLVM. i have to stop pushing to master

2023-05-26 03:11:03 +00:00

test_speed_v_torch.py

global -> group (#1007)

2023-06-21 11:50:43 -07:00

test_tensor.py

Zero dim Tensor support (#777)

2023-06-01 11:32:02 -07:00