diff options
author | Johan Alfven <johan.alfven@arm.com> | 2024-04-20 08:04:27 +0200 |
---|---|---|
committer | Johan Alfven <johan.alfven@arm.com> | 2024-04-24 12:16:57 +0200 |
commit | f49370003956d4f6f7d177114a68edb07b364fe9 (patch) | |
tree | fcd614f6f8293a44e789b723b5dee663aa3957c1 /ethosu/vela/tosa_reader.py | |
parent | f9194e11b2d2f1e813e347cad2d863a26115fa7e (diff) | |
download | ethos-u-vela-f49370003956d4f6f7d177114a68edb07b364fe9.tar.gz |
MLBEDSW-8969: Enable weight buffering for fully connected with batch shape
- Fully connected with batch shape will use the weights
more than once. Models with these type of fully connected
will benefit from weight buffering.
- If a fully connected op with this shape is detected it is
changed to a conv2d and the normal weight buffering
flow will be used.
Change-Id: I272741a32390e036d5e04bd5af41d4538162e86e
Signed-off-by: Johan Alfven <johan.alfven@arm.com>
Diffstat (limited to 'ethosu/vela/tosa_reader.py')
0 files changed, 0 insertions, 0 deletions