George HotzandGitHub 386fb5e7f8 folding without UNMUL (#5628) ...

* folding without UNMUL

* fix failures, index_collapse

* import ReduceOps

* test_arange_4096 isn't folding

2024-07-21 20:14:44 -07:00

..

codegen

folding without UNMUL (#5628)

2024-07-21 20:14:44 -07:00

engine

mcts exit condition wasn't right, also use it with BEAM>=100 (#5619)

2024-07-21 10:16:47 -07:00

nn

uniform init to match torch (#5494)

2024-07-15 12:07:44 -07:00

renderer

move UPat and PatternMatcher from uopgraph.py to uops.py (#5597)

2024-07-19 19:28:24 -04:00

runtime

helpers: remove duplicate data64 helpers in amd/nv (#5627)

2024-07-21 16:50:59 -07:00

shape

use exec_alu in uops flop counting (#5511)

2024-07-17 22:39:27 +03:00

__init__.py

add mnist tutorial

2024-04-25 16:08:32 +08:00

device.py

hcq move map to allocator (#5610)

2024-07-20 19:02:45 +03:00

dtype.py

pretty print lazy op per default (#5505)

2024-07-18 09:34:08 -07:00

function.py

threefry as pattern matcher (#5371)

2024-07-13 11:59:03 -07:00

helpers.py

helpers: remove duplicate data64 helpers in amd/nv (#5627)

2024-07-21 16:50:59 -07:00

lazy.py

utility for computing reduceop output_shape (#5534)

2024-07-17 22:40:07 +03:00

multi.py

s/loadops/metaops [run_process_replay] (#5421)

2024-07-12 13:26:50 -07:00

ops.py

fix no locals behavior (#5593)

2024-07-19 14:35:09 -07:00

tensor.py

touchup Tensor.interpolate (#5525)

2024-07-17 13:35:57 -04:00