Juarez Bochi 18d30005c5 Add support to UL2 model family (#1300) ...

* Add support to UL2 model family

* Update docs with UL2

* Create ActivationWithOptionalGating to avoid polluting activations

* Also refactor quantized t5

* Remove useless conversion

* Revert Activation::NewGelu name change

* Remove useless return

* Apply rustfmt and clippy recommendations

* Reuse t5::ActivationWithOptionalGating in quantized version

* (cosmetic change) use a match rather than ifs + avoid early returns.

---------

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

2023-11-09 18:55:09 +01:00

..

cpu

Add an erf based gelu op (#900)

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

quantized

Allow using gguf-v3 files. (#1262)

2023-11-03 23:07:53 +01:00

accelerate.rs

Accelerate support for gelu. (#782)

2023-09-08 21:58:56 +01:00

backend.rs

Add the conv-transpose1d op. (#1251)

2023-11-03 09:44:46 +01:00

backprop.rs

Detach all grads during backprop. (#1243)

2023-11-05 14:07:41 +01:00

conv.rs

Test for the transposed conv1d. (#1254)

2023-11-03 13:10:28 +01:00

convert.rs

Add support for i64 (#563)

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

cpu_backend.rs

Add the conv-transpose1d op. (#1251)

2023-11-03 09:44:46 +01:00

cuda_backend.rs

Add the conv-transpose1d op. (#1251)

2023-11-03 09:44:46 +01:00

cudnn.rs

Override the default cudnn heuristics. (#957)

2023-09-25 10:31:53 +01:00

device.rs

Add a hack for generating random uniform/normal for f16/bf16. (#1228)

2023-10-31 20:27:59 +00:00

display.rs

Fix the debug implementation. (#648)

2023-08-28 22:51:39 +01:00

dtype.rs

Only optimize float tensors. (#1069)

2023-10-10 09:05:41 +01:00

dummy_cuda_backend.rs

Add the conv-transpose1d op. (#1251)

2023-11-03 09:44:46 +01:00

error.rs

Implemented meshgrid (#1174)

2023-10-25 12:49:11 +01:00

indexer.rs

Tensor based indexing. (#842)

2023-09-14 07:47:07 +01:00

layout.rs

Retrieve more information from PyTorch checkpoints. (#515)

2023-08-19 15:05:34 +01:00

lib.rs

Allow for different behavior between training and eval (#1213)

2023-10-29 07:53:09 +01:00

mkl.rs

Add tanh. (#675)

2023-08-30 13:54:50 +01:00

npy.rs

Fix the npy read function and add some testing. (#1080)

2023-10-12 15:25:05 +02:00

op.rs

Add support to UL2 model family (#1300)

2023-11-09 18:55:09 +01:00

pickle.rs

Handle LongStorage in pytorch checkpoints. (#1152)

2023-10-22 18:34:36 +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

PyO3: Add equal and __richcmp__ to candle.Tensor (#1099)

2023-10-30 15:17:28 +00:00

storage.rs

Add the conv-transpose1d op. (#1251)

2023-11-03 09:44:46 +01:00

strided_index.rs

Broadcasting performance optimization (cpu) (#182)

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

tensor.rs

[ONNX] Support a couple more ops. (#1284)

2023-11-06 22:44:58 +01:00

test_utils.rs

Move the test-utils bits to a shared place. (#619)

2023-08-27 09:42:22 +01:00

utils.rs

Q6K quantization (#495)

2023-08-17 22:22:57 +01:00

variable.rs

Improve the mnist training example. (#276)

2023-07-29 16:28:22 +01:00