Francis Lam 46850a0269 search: add a BEAM_COMPARE env to optionally not compare to hc/tc (#4107) ...

* search: add a BEAM_COMPARE env to optionally not compare to hc/tc

setting BEAM_COMPARE=0 will prevent additional memory allocation
needed to do the timing tests assuming the BEAM result is in
the diskcache.

* change to optionally use Buffer.allocate

2024-04-08 18:54:01 -04:00

..

extract_policynet.py

move dtypes to dtype.py (#2964)

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

extract_sa_pairs.py

move dtypes to dtype.py (#2964)

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

generate_dataset.sh

update kernel dataset to exclude the disktensor ones (#3651)

2024-03-07 17:35:19 -05:00

get_action_space.py

multioutput uoping infra (#3706)

2024-03-15 21:56:59 -07:00

helpers.py

fuzz_linearizer: add --ast and --file params to read kernels (#3877)

2024-03-22 14:27:40 -04:00

pretrain_valuenet.py

move dtypes to dtype.py (#2964)

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

rl.py

fixed imports (#2185)

2023-10-30 22:07:17 -07:00

run_qnet.py

Q network for pruning BEAM / uops deduping / BEAM_ESTIMATE (#2142)

2023-10-27 10:53:06 -10:00

search.py

search: add a BEAM_COMPARE env to optionally not compare to hc/tc (#4107)

2024-04-08 18:54:01 -04:00

test_beam_search.py

use float32 for rand buffer in test_beam_search and test in metal (#3831)

2024-03-19 23:22:58 -04:00

test_net.py

fixed imports (#2185)

2023-10-30 22:07:17 -07:00

test_time_linearizer.py

KOPT is over, BEAM is upstream (#2071)

2023-10-16 09:46:03 -07:00