laurent b50bd880ce Only narrow when needed + deactivate the kv cache.

2023-06-29 19:07:52 +01:00

..

backprop.rs

Use broadcasted scalars for const tensors.

2023-06-29 11:56:40 +01:00

cpu_backend.rs

Put more requirements on the withdtype trait.

2023-06-29 11:37:42 +01:00

cuda_backend.rs

Use Map2 for binary ops.

2023-06-29 10:09:15 +01:00

device.rs

Refactor the hierarchy.

2023-06-27 11:57:27 +02:00

display.rs

Add some display tests + bugfixes.

2023-06-27 21:37:28 +01:00

dtype.rs

Put more requirements on the withdtype trait.

2023-06-29 11:37:42 +01:00

dummy_cuda_backend.rs

Simplify the pattern matching logic in the cuda backend.

2023-06-29 09:21:11 +01:00

error.rs

Only narrow when needed + deactivate the kv cache.

2023-06-29 19:07:52 +01:00

layout.rs

Get the cpu backend to compile.

2023-06-28 14:12:38 +01:00

lib.rs

Start refactoring the stride.

2023-06-28 12:57:30 +01:00

npy.rs

Refactor the hierarchy.

2023-06-27 11:57:27 +02:00

op.rs

Use Map2 for binary ops.

2023-06-29 10:09:15 +01:00

shape.rs

Use broadcasted scalars for const tensors.

2023-06-29 11:56:40 +01:00

storage.rs

Simplify the pattern matching logic in the cuda backend.

2023-06-29 09:21:11 +01:00

strided_index.rs

Get the cpu tests to run.

2023-06-28 14:32:02 +01:00

tensor.rs

Only narrow when needed + deactivate the kv cache.

2023-06-29 19:07:52 +01:00

utils.rs

PyTorch like display implementation.

2023-06-27 21:16:35 +01:00