Laurent Mazare 1c9e5394a5 Add a custom softmax implementation. (#744) ... * Add a custom softmax implementation. * Add softmaxlastdim to the benchmarks. * And add a test. * Support more dtypes. * Polish the code. * Use the slow implementation on cuda. * Add a todo for the cuda kernel.		2023-09-05 14:20:23 +01:00
..
bert	Upgrading hf-hub.	2023-08-29 14:18:54 +02:00
bigcode	Add a yolo-v3 example. (#528)	2023-08-20 18:19:37 +01:00
custom-ops	Remove some dead-code annotations. (#629)	2023-08-27 18:52:33 +01:00
dinov2	Move the imagenet specific bits to a separate file. (#571)	2023-08-23 16:42:09 +01:00
efficientnet	Use the hub weights for efficientnet. (#573)	2023-08-23 18:20:21 +01:00
falcon	Repeat-penalty in the falcon example. (#634)	2023-08-28 08:13:40 +01:00
llama	Add some optional repeat penalty. (#623)	2023-08-27 10:48:45 +01:00
llama2-c	Add a repeat penality to the llama2-c command line example. (#713)	2023-09-01 20:38:58 +01:00
llama_multiprocess	VarBuilder cleanup (#627)	2023-08-27 18:03:26 +01:00
mnist-training	Add the optimizer trait. (#702)	2023-09-01 12:55:39 +01:00
musicgen	Proper log buckets for t5. (#727)	2023-09-03 20:33:50 +01:00
quantized	Tweak some quantized args (#692)	2023-08-31 17:25:21 +01:00
stable-diffusion	Add a custom softmax implementation. (#744)	2023-09-05 14:20:23 +01:00
whisper	Whisper fix (#711)	2023-09-01 20:04:07 +01:00
yolo-v3	Dilated convolutions (#657)	2023-08-29 16:12:11 +01:00
yolo-v8	Simplify usage of the pool functions. (#662)	2023-08-29 19:12:16 +01:00