10417 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
George Hotz	861323c121	cleanup convdw torch	2022-06-11 08:32:57 -07:00
George Hotz	041ad55ed1	clean up torch processing op	2022-06-11 08:27:09 -07:00
George Hotz	fc7eabb86f	processing op	2022-06-11 08:12:02 -07:00
George Hotz	72186ebd5a	movement ops, reshape is a copy now	2022-06-10 20:01:47 -07:00
George Hotz	e5d694490f	reshape makes a copy	2022-06-10 19:49:04 -07:00
George Hotz	c8bacd0d8e	rename transpose to permute	2022-06-10 19:41:50 -07:00
George Hotz	462f1ce0da	Remove Matmul (#323)	2022-06-10 19:26:23 -07:00
George Hotz	66cb86c9ec	fix strided conv	2022-06-10 15:17:53 -07:00
George Hotz	8864b37333	fix torch convdw	2022-06-10 15:04:39 -07:00
George Hotz	aac1a9b419	this breaks tests	2022-06-10 12:20:42 -07:00
George Hotz	c4238187be	oops, we should have a linter in a precommit hook	2022-06-10 10:35:50 -07:00
George Hotz	160f651c25	conv_args not in llops interface	2022-06-10 10:35:10 -07:00
George Hotz	8c084b8c12	dispatcher: now this is nice	2022-06-09 10:28:46 -07:00
George Hotz	70bb3a7976	remove more weird logic in the dispatcher	2022-06-09 10:06:01 -07:00
George Hotz	e58d344759	simpler dispatch logic since the mlops are universal now	2022-06-09 10:01:10 -07:00
George Hotz	273af8d732	remove unused CPUBuffer functions	2022-06-09 09:45:39 -07:00
George Hotz	9d85067ad8	no splatting conv_args	2022-06-09 09:40:58 -07:00
George Hotz	259b536e3a	clean up reduce a little	2022-06-09 09:34:19 -07:00
George Hotz	d3e7238bdd	remove dead code	2022-06-09 09:25:40 -07:00
George Hotz	40f8eb7383	namedtuple for conv_args	2022-06-09 09:19:52 -07:00
George Hotz	d0c3204996	get_conv_args	2022-06-09 09:10:40 -07:00
George Hotz	6dcaee2361	ops enum cleanup	2022-06-09 08:42:42 -07:00
George Hotz	e01ed64d7c	restore that naming	2022-06-09 08:38:34 -07:00
George Hotz	e7ebbecf97	proper llops	2022-06-09 08:33:25 -07:00
George Hotz	d841fc4392	stupid, but the tests should pass	2022-06-08 23:43:10 -07:00
George Hotz	214fb8c974	ctx.op	2022-06-08 23:39:54 -07:00
George Hotz	5a533fc073	llops torch test passes	2022-06-08 23:30:56 -07:00
George Hotz	1e3db466cc	torch llops	2022-06-08 23:15:03 -07:00
George Hotz	30ab2249eb	match order	2022-06-08 11:46:51 -07:00
George Hotz	4a9882d495	hlops	2022-06-08 11:46:09 -07:00
George Hotz	e046a2fd9f	readme fix typos	2022-06-08 11:43:05 -07:00
George Hotz	4b09ca90a1	readme: still WIP	2022-06-08 11:41:19 -07:00
George Hotz	91d97c8313	mlops for cpu	2022-06-08 11:20:10 -07:00
George Hotz	1764a36bd6	llops cpu passes	2022-06-08 10:57:45 -07:00
George Hotz	81d16d105e	llcpu convs work	2022-06-08 10:51:09 -07:00
George Hotz	6bdcf5ef59	llops cpu	2022-06-08 10:43:02 -07:00
George Hotz	582ddef59c	14 llops too	2022-06-08 10:01:10 -07:00
George Hotz	60a48455ad	still over line count, maybe test pass	2022-06-08 09:51:28 -07:00
George Hotz	70561f3d90	way over the line limit	2022-06-08 09:36:31 -07:00
George Hotz	4f7ee235c5	not a real test now	2022-06-08 09:00:59 -07:00
George Hotz	ae33060dae	early float4 stuff for binary	2022-06-08 08:59:54 -07:00
George Hotz	82f29b5dbf	better GPU block	2022-06-08 08:01:04 -07:00
George Hotz	42ae78241e	only run test on GPU	2022-06-08 07:54:40 -07:00
George Hotz	cdf4b5f142	opencl perf test	2022-06-08 07:49:08 -07:00
George Hotz	892ac661e1	enet readability	2022-06-07 10:23:05 -07:00
George Hotz	9b27ba650b	load new torch files	2022-06-07 10:06:48 -07:00
George Hotz	33c9c6c025	float4 for unary_op	2022-06-06 10:40:30 -07:00
George Hotz	69a401d596	use cl.enqueue_copy and add padding to GPUBuffers	2022-06-06 10:27:08 -07:00
George Hotz	d8ee8a39ac	sgd threestep graph is so pretty	2022-06-06 09:45:37 -07:00
George Hotz	aa3fb10a92	hmm, not allowed	2022-06-06 09:40:15 -07:00