George HotzandGitHub 2a2d3233d2 add test that the compiler isn't used (#3025) ...

* add test that the compiler isn't used

* one print_tree

* improve speed with st size cache

* switch to gpt-2

2024-01-05 17:24:01 -08:00

..

codegen

minor cleanup of matvec in hand_coded_optimizations (#3015)

2024-01-04 19:43:49 -05:00

features

hotfix: make multitensor shard contiguous

2024-01-03 08:48:30 -08:00

nn

replace d[a] if a in d else b with d.get(a, b) (#2997)

2024-01-03 18:10:25 -05:00

renderer

Bitcast hip fix + fix mixtral (#3022)

2024-01-05 14:51:25 -08:00

runtime

fast hip read (#3014)

2024-01-05 10:33:13 -08:00

shape

add test that the compiler isn't used (#3025)

2024-01-05 17:24:01 -08:00

__init__.py

move dtypes to dtype.py (#2964)

2024-01-01 14:58:48 -08:00

device.py

add test that the compiler isn't used (#3025)

2024-01-05 17:24:01 -08:00

dtype.py

pass tuple for strs to startswith (#2986)

2024-01-02 19:51:15 -05:00

graph.py

hotfix: save lines in graph

2024-01-03 12:03:42 -08:00

helpers.py

add test that the compiler isn't used (#3025)

2024-01-05 17:24:01 -08:00

jit.py

hotfix: don't jitcache with 1 kernel. improvements to hip sniffer

2024-01-03 19:17:08 -08:00

lazy.py

Bitcast hip fix + fix mixtral (#3022)

2024-01-05 14:51:25 -08:00

mlops.py

grad -> grad_output in mlops for consistency (#2982)

2024-01-02 18:03:55 -05:00

ops.py

vars_from_ast -> LazyOp.vars (#2965)

2024-01-01 18:12:38 -05:00

realize.py

don't alloc for InterpretedASTRunner (#2999)

2024-01-03 17:05:53 -08:00

tensor.py

add Tensor.unflatten and Tensor.flatten(end_dim) (#3023)

2024-01-05 17:55:29 -05:00