wandering007 / nasnet-pytorch

A neat pytorch implementation of nasnet and the ported weights from tensorflow

A neat pytorch implementation of NASNet

The performance of the ported models on ImageNet (Accuracy):

Model Checkpoint	Million Parameters	Val Top-1	Val Top-5
NASNet-A_Mobile_224	5.3	70.2	89.4
NASNet-A_large_331	88.9	82.3	96.0

The slight performance drop may be caused by the different spatial padding methods between tensorflow and pytorch.

The porting process is done by tensorflow_dump.py and pytorch_load.py, modified from Cadene's project. Note that NASNets with the original performance can be found there.

You can evaluate the models by running imagenet_eval.py, e.g. evaluate the NASNet-A_Mobile_224 ported model by

python imagenet_eval.py --nas-type mobile --resume /path/to/modelfile --gpus 0 --data /path/to/imagenet_root_dir

The ported model files are provided: NASNet-A_Mobile_224, NASNet-A_large_331.

Future work:

• add drop path for training
• more nasnet model settings