Skip to content

Quantization: MLP calibration, GNN export (need for TensorRT INT8 plugin)

Fotis Giasemis requested to merge fgiasemi/quantization into main
  • Calibrate MLP
  • Export quantized GNN (?) --> Need to write TensorRT plugin for this

To do:

  • New MR for quantization of only the MLPs of the GNN
Edited by Fotis Giasemis

Merge request reports