Closed Aspirinkb closed 5 years ago
This is outside of the scope of gemmlowp, which is only a matrix multiplication library. Here is how addition of quantized arrays is implemented in TFLite - reference (slow portable) code here: https://github.com/tensorflow/tensorflow/blob/458e0e300bbb8e9c9824a960fef13877983369f4/tensorflow/lite/kernels/internal/reference/reference_ops.h#L566-L597 more optimized NEON code in ../optimized/optimized_ops.h.
Adding two feature maps in ResNet is a common operation, but how to add outputs of two layers with different
scale
s andzero_point
s?Let's say
r3 = r1*r2
, andr1 = s1*(q1-z1)
,r2 = s2*(q2-z2)
r3 = s3 * (q3-z3)
.So how to get
q3
?Obviously,
q3 = t1*(q1-z1) + t2*(q2-z2) + z3
, wheret1 = s1/s3
t2 = s2/s3
.How to get the result of
t1*(q1-z1)
?