projects / platform / upstream / armnn.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: fab64f0) | patch
IVGCVSW-2436 Modify MobileNet SSD inference test
authorNarumol Prangnawarat <narumol.prangnawarat@arm.com>
Fri, 15 Feb 2019 17:34:51 +0000 (17:34 +0000)
committerNarumol Prangnawarat <narumol.prangnawarat@arm.com>
Wed, 20 Feb 2019 11:06:06 +0000 (11:06 +0000)
commitc8bab1bf601b00242b44cdbc7d9510a753054803
tree05fa368fac3417cc30f34b791d165879f81ba1e8tree | snapshot
parentfab64f09f49280ce7e57837c8bc99de31089512ccommit | diff
IVGCVSW-2436 Modify MobileNet SSD inference test

 * change MobileNet SSD input to uint8
 * get quantization scale and offset from the model
 * change data layout to NHWC as TensorFlow lite layout
 * update expected output as result from TfLite with quantized data

Change-Id: I07104d56286893935779169356234de53f1c9492
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
tests/MobileNetSsdDatabase.hpp diff | blob | history
tests/MobileNetSsdInferenceTest.hpp diff | blob | history
Domain: Machine Learning / ML Framework;