README.md

1. Training MobileNet on CIFAR-10.

   Training:

   python train.py data/ --dir models/mobilenet/model.pth.tar --arch mobilenet

   Evaluation:

   python eval.py data/ --dir models/mobilenet/model.pth.tar --arch mobilenet

   One trained model can be found here.

2. Measuring Latency

   Here we build the latency lookup table for cuda:0 device:

   python build_lookup_table.py --dir latency_lut/lut_mobilenet.pkl --arch mobilenet

   It measures latency of different layers contained in the network (i.e. MobileNet here). For conv layers, the sampled numbers of feature channels are multiples of MIN_CONV_FEATURE_SIZE. For fc layers, the sampled numbers of features are multiples of MIN_FC_FEATURE_SIZE.

3. Applying NetAdapt

   Modify which GPUs will be utilized (-gp) in netadapt_mobilenet-0.5mac.sh and run the script to apply NetAdapt to a pretrained model:

   sh scripts/netadapt_mobilenet-0.5mac.sh

   You can see how the model is simplified at each iteration in models/mobilenet/prune-by-mac/master/history.txt and select the one that satisfies the constraints to run long-term fine-tune.

   After obtaining the adapted model, we need to finetune the model (here we select the one after 28 iterations):

   python train.py data/ --arch mobilenet --resume models/mobilenet/prune-by-mac/master/iter_28_best_model.pth.tar --dir models/mobilenet/prune-by-mac/master/finetune_model.pth.tar --lr 0.001

   

   If you want to get a model with 50% latency, please run:

   sh scripts/netadapt_mobilenet-0.5latency.sh

4. Evaluation Using Adapted Models

   After applying NetAdapt to a pretrained model, we can evaluate this adapted model using:

   python eval.py data/ --dir models/mobilenet/prune-by-mac/master/finetune_model.pth.tar --arch mobilenet

   The adapted model can be restored without modifying the orignal python file.

   We provide one adapted model here.