The following are some tips for quantize finetuning.
- Dropout: Experiments shows that quantize finetuning works better without dropout ops. This tool does not support quantize finetuning with dropouts now, they should be removed or disabled before running the quantize finetuning. This can be done by setting is_training=false when using tf.layers or call tf.keras.backend.set_learning_phase(0) when using tf.keras.layers.
- Hyper-param: Quantize finetuning is like float finetuning, so the techniques for float finetuning is also needed. The optimizer type, learning rate curve are some important parameters to tune.