George Hotz 643cbdfd50 make embedding and GPT-2 fast (#1631) ...

* make embedding fast

* jit more, variable shape support

* print mem bw

2023-08-22 15:14:38 -07:00

..

assembly_arm64.py

refactor: no len if it is not needed (#1598)

2023-08-21 14:06:32 -07:00

assembly_ptx.py

PTX Reintegration and Passing Tests (#1512)

2023-08-16 16:20:20 -07:00

assembly.py

refactor: no len if it is not needed (#1598)

2023-08-21 14:06:32 -07:00

linearizer.py

make embedding and GPT-2 fast (#1631)

2023-08-22 15:14:38 -07:00

optimizer.py

make embedding and GPT-2 fast (#1631)

2023-08-22 15:14:38 -07:00