CSMPQ

Requirements

• DALI (for accelerating data processing)
• Apex (for distributed running)
• other requirements, running requirements.txt

pip install -r requirements.txt

Running

Bit Configuration

#!/usr/bin/env bash
python3 -m torch.distributed.launch --nproc_per_node=1 feature_extract_cdp.py \
 --model "resnet18" \
 --path "/Path/to/Base_model" \      # pretrained base model
 --dataset "imagenet" \
 --save_path '/Path/to/Dataset/' \   # Dataset path
 --beta 10.0 \                       # Hyper-parameter for bit difference
 --model_size 6.7 \                  # Target model size
 --quant_type "QAT"                  # Post-Training Quantization(PTQ) or Quantization-Aware Training(QAT)

or

bash ./mixed_bit/run_scripts/QAT/quant_resnet18.sh

QAT

Because of random seed, bit configuration obtained through feature extraction may have a little difference from ours. Our bit configurations are given in bit_config.py. Our quantized models and logs are also given in this link.

#!/usr/bin/env bash
python quant_train.py \
 -a resnet18 \
 --epochs 90 \
 --lr 0.0001 \
 --batch_size 128 \
 --data /Path/to/Dataset/ \
 --save_path /Path/to/Save_quant_model/ \
 --act_range_momentum=0.99 \
 --wd 1e-4 \
 --data_percentage 1 \
 --pretrained \
 --fix_BN \
 --checkpoint_iter -1 \
 --quant_scheme modelsize_6.7_a6_75B

or

bash ./QAT/run_scripts/train_resnet18.sh

PTQ

For the post-training quantization, we only require a few GPU hours to get the quantization model. So we set the random seed. You can directly get the same accuracy in the paper by running codes as follows:

python main_imagenet.py --data_path /Path/to/Dataset/ --arch resnet18 --n_bits_w 2 --channel_wise --n_bits_a 8 --act_quant --test_before_calibration --bit_cfg "[4, 3, 3, 4, 4, 4, 4, 4, 4, 4, 3, 3, 4, 4, 3, 3, 3, 3]"

or

bash ./PTQ/run_scripts/train_resnet18.sh

Related Works

• BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstruction (ICLR 2021)