INT4 and other low-precision conversion support status #64193
Labels
comp:lite
TF Lite related issues
ModelOptimizationToolkit
TF Model Optimization Toolkit
stat:awaiting tensorflower
Status - Awaiting response from tensorflower
TFLiteConverter
For issues related to TFLite converter
type:feature
Feature requests
What is the current status of model conversion (PTQ specifically) with INT4 precision?
The question was raised before here #60125.
Also it looks like INT4 support is being added to various parts of tensorflow, as evidenced by #63870 and
tensorflow/tensorflow/compiler/mlir/quantization/tensorflow/quantization_options.proto
Line 76 in 6738c28
However at the moment there seems to be no way to quantize model to INT4 (specifically the weights):
Can anyone who actively works on this in TF team shine the light on what is the current direction and where one needs to dig to add INT4 PTQ quantization?
The text was updated successfully, but these errors were encountered: