diff options
author | Dwight Lidman <dwight.lidman@arm.com> | 2021-03-15 19:06:10 +0100 |
---|---|---|
committer | patrik.gustavsson <patrik.gustavsson@arm.com> | 2021-03-16 16:29:01 +0000 |
commit | 9b3791817c32529dfeddae57c29c2abe19311fc4 (patch) | |
tree | 866c769bb8627b686f7d25ddce3413380c34cbdb /SUPPORTED_OPS.md | |
parent | c822d62ba27b874a130e9d8d434c12b419d10d62 (diff) | |
download | ethos-u-vela-9b3791817c32529dfeddae57c29c2abe19311fc4.tar.gz |
MLBEDSW-4215: Add support for MEAN to match QuantizedMeanOrSum implementation
This commit adds support for emulating the behavior
of the QuantizedMeanOrSum implementation of MEAN in
TensorFlow Lite.
Signed-off-by: Dwight Lidman <dwight.lidman@arm.com>
Change-Id: Ifd24e0e678e2f85cd66ab82deeaaf010d5351b1e
Diffstat (limited to 'SUPPORTED_OPS.md')
-rw-r--r-- | SUPPORTED_OPS.md | 8 |
1 files changed, 0 insertions, 8 deletions
diff --git a/SUPPORTED_OPS.md b/SUPPORTED_OPS.md index 3c90e20c..1ad65c63 100644 --- a/SUPPORTED_OPS.md +++ b/SUPPORTED_OPS.md @@ -197,14 +197,6 @@ This is a list of constraints that the MAX_POOL_2D operator must satisfy in orde This is a list of constraints that the MEAN operator must satisfy in order to be scheduled on the NPU. - IFM must be int8 or uint8 -- Every constraint in either one (or both) of the following sets of constraints must be fulfilled: - Set A: - IFM dimensions are 4, - Axis indices are 1 and 2, - keep_dims is set to True - Set B: - IFM zero point and OFM zero point are the same, - IFM scale and OFM scale are the same - Input tensor must be at least 2D - Axis indices must correspond to height and width axes - Product of height and width can be at most 4096 |