diff options
| author | Andreas Nevalainen <andreas.nevalainen@arm.com> | 2020-11-16 09:14:52 +0100 |
|---|---|---|
| committer | patrik.gustavsson <patrik.gustavsson@arm.com> | 2020-11-17 07:59:59 +0000 |
| commit | 15a8e803844b286fe9533e1cf703c76a77b090a8 (patch) | |
| tree | b7ccbc219579c80babbce022ea0e52733ed228b2 /ethosu/vela/mark_tensors.py | |
| parent | c86a9d2fe531dee57886b7adf6eb16dc16132650 (diff) | |
| download | ethos-u-vela-15a8e803844b286fe9533e1cf703c76a77b090a8.tar.gz | |
MLMBED-3450: Do not convert batched fully connected to conv
Do not convert batched fully connected operators to avoid moving
weights from flash to SRAM.
Change-Id: I873c9ce05377de3f16e4cee9a0863f29d9ec3ad4
Signed-off-by: Andreas Nevalainen <andreas.nevalainen@arm.com>
Diffstat (limited to 'ethosu/vela/mark_tensors.py')
0 files changed, 0 insertions, 0 deletions