The implementation of this proposal has been submit to
https://github.com/apache/incubator-tvm/pull/4459 .
Anyone can try to test their TVM operators by re-compiling TVM with
`set(USE_TFOP ON)`.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly o
## Problem
TensorFlow is one of the most popular machine learning libraries and most
developers are used to train/inference models with TensorFlow/TensorFlow
Serving. TVM is the flexible compiler to run computation efficiently in
different devices. Although TensorFlow has implemented some effic
agree with tqchen. The difference between fixed point and integer is that fixed
point has fractional part while integer does not. By adding "point position
(pp)" and adding dtype to quantization pass, we could reuse most of
quantization API for the same matter. Of course, the TVM compute needs t
+1
--
You are receiving this because you commented.
Reply to this email directly or view it on GitHub:
https://github.com/apache/incubator-tvm/issues/4443#issuecomment-561196336
