MilkFather 3bbb88fcb4 Fix sigmoid gradient calculation and move sigmoid into a specialized op (#2114) ...

* add sigmoid op

* small fix

* add as a method on `Tensor`

* implement gradient calculation for sigmoid

* add sigmoid tests

* we should have a specialized op for this

* fix clippy

* fix clippy 2

* Revert all previous commits in favor of a `CustomOp` based solution

* use `CustomOp1` implementation

* fix rustfmt

* experimental add metal impl

* add cuda kernel impl

* fix fmt

* Add a test + reduce some cuda duplication.

---------

Co-authored-by: laurent <laurent.mazare@gmail.com>

2024-04-29 11:04:43 +02:00

..

cpu

Add an erf based gelu op (#900)

2023-09-19 19:54:28 +01:00

cpu_backend

Add StorageRef. (#2113)

2024-04-23 13:23:27 +02:00

cuda_backend

Fix sigmoid gradient calculation and move sigmoid into a specialized op (#2114)

2024-04-29 11:04:43 +02:00

metal_backend

Add argsort. (#2132)

2024-04-27 20:17:35 +02:00

quantized

Add a forward_via_f16 method to the qmatmul op. (#2138)

2024-04-28 20:35:01 +02:00

accelerate.rs

feat: add silu activation function (#1706)

2024-02-14 10:27:22 +01:00

backend.rs

Add StorageRef. (#2113)

2024-04-23 13:23:27 +02:00

backprop.rs

Fix the silu gradient issue on 0. (#2083)

2024-04-18 14:31:41 +02:00

conv.rs

Support for groups in conv-transpose1d. (#1731)

2024-02-18 21:28:07 +01:00

convert.rs

Add support for i64 (#563)

2023-08-23 10:42:19 +01:00

custom_op.rs

Preliminary support for inplace ops. (#1921)

2024-03-23 14:16:19 +01:00

device.rs

Add StorageRef. (#2113)

2024-04-23 13:23:27 +02:00

display.rs

Expose more printer options. (#1817)

2024-03-08 15:04:18 +01:00

dtype.rs

Add StorageRef. (#2113)

2024-04-23 13:23:27 +02:00

dummy_cuda_backend.rs

Add a toggle for F16/BF16 accumulation in gemm. (#2141)

2024-04-29 09:21:07 +02:00

dummy_metal_backend.rs

Add StorageRef. (#2113)

2024-04-23 13:23:27 +02:00

error.rs

Small cleanups to the llama multi-process example. (#2098)

2024-04-20 22:19:46 +02:00

indexer.rs

Fix a couple typos (#1451)

2023-12-17 05:20:05 -06:00

layout.rs

Expose a couple layout methods. (#1816)

2024-03-08 10:52:22 +01:00

lib.rs

Add a toggle for F16/BF16 accumulation in gemm. (#2141)

2024-04-29 09:21:07 +02:00

mkl.rs

feat: add silu activation function (#1706)

2024-02-14 10:27:22 +01:00

npy.rs

Update zip requirement from 0.6.6 to 1.1.1 (#2103)

2024-04-22 16:23:27 +02:00

op.rs

Add support for "sign" on tensors (#2012)

2024-04-04 22:32:47 +02:00

pickle.rs

Fix float unpickling. (#1730)

2024-02-18 19:33:55 +01:00

safetensors.rs

Add the buffered safetensor wrapper. (#948)

2023-09-23 22:57:42 +01:00

scalar.rs

Get the comparison operation to work on scalar values. (#780)

2023-09-08 20:13:29 +01:00

shape.rs

Another fix for squeezing. (#1943)

2024-03-26 17:05:26 +01:00

sort.rs

Add a sort function. (#2134)

2024-04-28 08:18:04 +02:00

storage.rs

Really unique identifier for metal device ids. (#1932)

2024-03-25 11:48:16 +01:00

strided_index.rs

Broadcasting performance optimization (cpu) (#182)

2023-07-17 13:41:09 +01:00

tensor_cat.rs

Remove some unnecessary calls to contiguous. (#1968)

2024-03-30 13:22:00 +01:00

tensor.rs

Add StorageRef. (#2113)

2024-04-23 13:23:27 +02:00

test_utils.rs

Adding the test scaffolding.

2023-11-20 14:38:35 +01:00

utils.rs

Metal part 1 - Scaffolding for metal. (#1308)

2023-11-10 08:35:48 +01:00

variable.rs

Detach the tensors on batch-norm eval. (#1702)

2024-02-13 14:26:32 +01:00