tvm-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Wuwei Lin via TVM Discuss <>
Subject [TVM Discuss] [Development] Improving quantization accuracy with more precise bias
Date Wed, 08 May 2019 17:38:11 GMT

The above example after annotation:
|                            |
sim_quantize(QINPUT) sim_quantize(QINPUT)
|                            |
...                     / 
data is usually output of previous conv2d. There are duplicated simulated_quantize. Followed
add in both branches will convert the int8 to int32. So simulated_quantize + add in both branches
which will be translated to `right_shift + cast(i8) + cast(i32)`
We use stop_fusion to ensure that previous conv2d result will be casted to int8 before saving
in global memory.

You will see the difference running quantized ResNet-50 v2.

[Visit Topic](
to respond.

You are receiving this because you enabled mailing list mode.

To unsubscribe from these emails, [click here](

Tianqi Chen, UW, Seattle, WA, 98105, United States
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message