update codes

README.md

@@ -6,6 +6,7 @@ University of Technology Sydney
 Requirements
 - PyTorch 1.0
 - Python 3.6
+- opencv
 ```
 conda install pytorch torchvision cuda100 -c pytorch
 ```
@@ -14,12 +15,13 @@ conda install pytorch torchvision cuda100 -c pytorch
 
 Searching CNNs
 ```
+bash ./scripts-cnn/search-acc-v2.sh 3 acc2
 ```
 
 Train the searched CNN on CIFAR
 ```
-bash ./scripts-cnn/train-imagenet.sh 0 GDAS_F1 52 14
+bash ./scripts-cnn/train-cifar.sh 0 GDAS_F1 cifar10
-bash ./scripts-cnn/train-imagenet.sh 0 GDAS_V1 50 14
+bash ./scripts-cnn/train-cifar.sh 0 GDAS_V1 cifar100
 ```
 
 Train the searched CNN on ImageNet

configs/nas-cifar-cos-cut-v1.config

@@ -1,13 +0,0 @@
-{
-  "type"      : ["str",   "cosine"],
-  "batch_size": ["int",   128],
-  "epochs"    : ["int",   600],
-  "momentum"  : ["float", 0.9],
-  "decay"     : ["float", 0.0005],
-  "LR"        : ["float", 0.025],
-  "auxiliary" : ["bool", 1],
-  "auxiliary_weight" : ["float", 0.4],
-  "grad_clip" : ["float", 5],
-  "cutout"    : ["int", 16],
-  "drop_path_prob" : ["float", 0.2]
-}

exps-cnn/train_base.py

@@ -54,15 +54,15 @@ def main():
   if not os.path.isdir(args.save_path):
     os.makedirs(args.save_path)
   log = open(os.path.join(args.save_path, 'log-seed-{:}.txt'.format(args.manualSeed)), 'w')
-  print_log('save path : {}'.format(args.save_path), log)
+  print_log('Save Path      : {:}'.format(args.save_path), log)
   state = {k: v for k, v in args._get_kwargs()}
   print_log(state, log)
-  print_log("Random Seed: {}".format(args.manualSeed), log)
+  print_log("Random Seed    : {:}".format(args.manualSeed), log)
-  print_log("Python version : {}".format(sys.version.replace('\n', ' ')), log)
+  print_log("Python version : {:}".format(sys.version.replace('\n', ' ')), log)
-  print_log("Torch  version : {}".format(torch.__version__), log)
+  print_log("Torch  version : {:}".format(torch.__version__), log)
-  print_log("CUDA   version : {}".format(torch.version.cuda), log)
+  print_log("CUDA   version : {:}".format(torch.version.cuda), log)
-  print_log("cuDNN  version : {}".format(cudnn.version()), log)
+  print_log("cuDNN  version : {:}".format(cudnn.version()), log)
-  print_log("Num of GPUs    : {}".format(torch.cuda.device_count()), log)
+  print_log("Num of GPUs    : {:}".format(torch.cuda.device_count()), log)
   args.dataset = args.dataset.lower()
 
   config = load_config(args.model_config)

exps-cnn/train_utils.py

@@ -21,7 +21,7 @@ def obtain_best(accuracies):
 
 def main_procedure(config, dataset, data_path, args, genotype, init_channels, layers, log):
   
-  train_data, test_data, class_num = get_datasets(dataset, data_path, args.cutout)
+  train_data, test_data, class_num = get_datasets(dataset, data_path, config.cutout)
 
   print_log('-------------------------------------- main-procedure', log)
   print_log('config        : {:}'.format(config), log)
@@ -39,9 +39,9 @@ def main_procedure(config, dataset, data_path, args, genotype, init_channels, la
   print_log('genotype      : {:}'.format(genotype), log)
   print_log('args          : {:}'.format(args), log)
   print_log('Train-Dataset : {:}'.format(train_data), log)
-  print_log('Train-Trans   : {:}'.format(train_transform), log)
+  print_log('Train-Trans   : {:}'.format(train_data.transform), log)
   print_log('Test--Dataset : {:}'.format(test_data ), log)
-  print_log('Test--Trans   : {:}'.format(test_transform ), log)
+  print_log('Test--Trans   : {:}'.format(test_data.transform ), log)
 
 
   train_loader = torch.utils.data.DataLoader(train_data, batch_size=config.batch_size, shuffle=True,

exps-cnn/train_utils_imagenet.py

@@ -62,7 +62,7 @@ def main_procedure_imagenet(config, data_path, args, genotype, init_channels, la
 
   total_param, aux_param = count_parameters_in_MB(basemodel), count_parameters_in_MB(basemodel.auxiliary_param())
   print_log('Network =>\n{:}'.format(basemodel), log)
-  #print_FLOPs(basemodel, (1,3,224,224), [print_log, log])
+  print_FLOPs(basemodel, (1,3,224,224), [print_log, log])
   print_log('Parameters : {:} - {:} = {:.3f} MB'.format(total_param, aux_param, total_param - aux_param), log)
   print_log('config        : {:}'.format(config), log)
   print_log('genotype      : {:}'.format(genotype), log)
@@ -75,7 +75,7 @@ def main_procedure_imagenet(config, data_path, args, genotype, init_channels, la
   criterion_smooth = CrossEntropyLabelSmooth(class_num, config.label_smooth).cuda()
 
 
-  optimizer = torch.optim.SGD(model.parameters(), config.LR, momentum=config.momentum, weight_decay=config.decay, nestero=True)
+  optimizer = torch.optim.SGD(model.parameters(), config.LR, momentum=config.momentum, weight_decay=config.decay, nesterov=True)
   if config.type == 'cosine':
     scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, float(config.epochs))
   elif config.type == 'steplr':

lib/datasets/get_dataset_with_transform.py

@@ -10,7 +10,7 @@ from .TieredImageNet import TieredImageNet
 Dataset2Class = {'cifar10' : 10,
                  'cifar100': 100,
                  'tiered'  : -1,
-                 'imagnet-1k'  : 1000,
+                 'imagenet-1k' : 1000,
                  'imagenet-100': 100}
 
 
@@ -25,8 +25,8 @@ def get_datasets(name, root, cutout):
     std = [x / 255 for x in [68.2, 65.4, 70.4]]
   elif name == 'tiered':
     mean, std = [0.485, 0.456, 0.406], [0.229, 0.224, 0.225]
-  elif name == 'imagnet-1k' or name == 'imagenet-100':
+  elif name == 'imagenet-1k' or name == 'imagenet-100':
-    mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]
+    mean, std = [0.485, 0.456, 0.406], [0.229, 0.224, 0.225]
   else: raise TypeError("Unknow dataset : {:}".format(name))
 
 
@@ -42,7 +42,7 @@ def get_datasets(name, root, cutout):
     if cutout > 0 : lists += [Cutout(cutout)]
     train_transform = transforms.Compose(lists)
     test_transform  = transforms.Compose([transforms.CenterCrop(80), transforms.ToTensor(), transforms.Normalize(mean, std)])
-  elif name == 'imagnet-1k' or name == 'imagenet-100':
+  elif name == 'imagenet-1k' or name == 'imagenet-100':
     normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
     train_transform = transforms.Compose([
       transforms.RandomResizedCrop(224),
@@ -57,15 +57,14 @@ def get_datasets(name, root, cutout):
     ])
     test_transform = transforms.Compose([transforms.Resize(256), transforms.CenterCrop(224), transforms.ToTensor(), normalize])
   else: raise TypeError("Unknow dataset : {:}".format(name))
-    train_data = TieredImageNet(root, 'train-val', train_transform)
+
-    test_data = None
   if name == 'cifar10':
     train_data = dset.CIFAR10(root, train=True, transform=train_transform, download=True)
     test_data  = dset.CIFAR10(root, train=True, transform=test_transform , download=True)
   elif name == 'cifar100':
     train_data = dset.CIFAR100(root, train=True, transform=train_transform, download=True)
     test_data  = dset.CIFAR100(root, train=True, transform=test_transform , download=True)
-  elif name == 'imagnet-1k' or name == 'imagenet-100':
+  elif name == 'imagenet-1k' or name == 'imagenet-100':
     train_data = dset.ImageFolder(osp.join(root, 'train'), train_transform)
     test_data  = dset.ImageFolder(osp.join(root, 'val'), train_transform)
   else: raise TypeError("Unknow dataset : {:}".format(name))

lib/utils/__init__.py

@@ -4,7 +4,6 @@ from .utils import test_imagenet_data
 from .utils import print_log
 from .evaluation_utils import obtain_accuracy
 from .draw_pts import draw_points
-from .fb_transform import ApplyOffset
 from .gpu_manager import GPUManager
 
 from .save_meta import Save_Meta

lib/utils/fb_transform.py

@@ -1,14 +0,0 @@
-import torch
-import random
-import numpy as np
-
-class ApplyOffset(object):
-  def __init__(self, offset):
-    assert isinstance(offset, int), 'The offset is not right : {}'.format(offset)
-    self.offset = offset
-  def __call__(self, x):
-    if isinstance(x, np.ndarray) and x.dtype == 'uint8':
-      x = x.astype(int)
-    if isinstance(x, np.ndarray) and x.size == 1:
-      x = int(x)
-    return x + self.offset

scripts-cnn/search-acc-v2.sh

@@ -0,0 +1,29 @@
+#!/usr/bin/env sh
+if [ "$#" -ne 2 ] ;then
+  echo "Input illegal number of parameters " $#
+  echo "Need 2 parameters for the GPUs and the network"
+  exit 1               
+fi 
+if [ "$TORCH_HOME" = "" ]; then
+  echo "Must set TORCH_HOME envoriment variable for data dir saving"
+  exit 1
+else
+  echo "TORCH_HOME : $TORCH_HOME"
+fi
+
+gpus=$1
+arch=$2
+cutout=0
+dataset=cifar10
+epoch=200
+SAVED=./snapshots/NAS/ACC-V2-Search-${arch}-${dataset}-cut${cutout}-${epoch}-E600
+
+CUDA_VISIBLE_DEVICES=${gpus} python ./exps-cnn/DARTS-Search.py \
+	--data_path $TORCH_HOME/cifar.python \
+	--arch ${arch} --dataset ${dataset} --batch_size 128 \
+	--save_path ${SAVED} \
+	--learning_rate_max 0.025 --learning_rate_min 0.001 --momentum 0.9 --weight_decay 0.0003 \
+	--epochs ${epoch} --cutout ${cutout} --validate --grad_clip 5 \
+	--init_channels 16 --layers 8 \
+	--model_config ./configs/nas-cifar-cos.config \
+	--print_freq 100 --workers 10

scripts-cnn/search.sh

@@ -0,0 +1,45 @@
+#!/usr/bin/env sh
+if [ "$#" -ne 3 ] ;then
+  echo "Input illegal number of parameters " $#
+  echo "Need 3 parameters for the GPUs and the network and the dataset"
+  exit 1               
+fi 
+if [ "$TORCH_HOME" = "" ]; then
+  echo "Must set TORCH_HOME envoriment variable for data dir saving"
+  exit 1
+else
+  echo "TORCH_HOME : $TORCH_HOME"
+fi
+
+gpus=$1
+arch=$2
+cutout=0
+dataset=$3
+epoch=50
+SAVED=./snapshots/NAS/Search-${arch}-${dataset}-cut${cutout}-${epoch}
+
+if [ "$dataset" == "cifar10" ] ;then
+  dataset_root=$TORCH_HOME/cifar.python
+  print_freq=100
+elif [ "$dataset" == "cifar100" ] ;then
+  dataset_root=$TORCH_HOME/cifar.python
+  print_freq=100
+elif [ "$dataset" == "tiered" ] ;then
+  dataset_root=$TORCH_HOME/tiered-imagenet
+  print_freq=500
+else
+  echo 'invalid dataset-name :'${dataset}
+  exit 1
+fi
+
+CUDA_VISIBLE_DEVICES=${gpus} python ./exps-cnn/DARTS-Search.py \
+	--data_path ${dataset_root} \
+	--arch ${arch} \
+	--dataset ${dataset} --batch_size 64 \
+	--save_path ${SAVED} \
+	--learning_rate_max 0.025 --learning_rate_min 0.001 --momentum 0.9 --weight_decay 0.0003 \
+	--epochs ${epoch} --cutout ${cutout} --validate --grad_clip 5 \
+	--init_channels 16 --layers 8 \
+	--manualSeed 3858 \
+	--model_config ./configs/nas-cifar-cos-cut.config \
+	--print_freq ${print_freq} --workers 8

scripts-cnn/train-cifar.sh

@@ -1,8 +1,8 @@
 #!/usr/bin/env sh
-# bash scripts-cnn/train-cifar.sh 0 GDAS cifar10
+# bash scripts-cnn/train-cifar.sh 0 GDAS cifar10 cut
-if [ "$#" -ne 3 ] ;then
+if [ "$#" -ne 4 ] ;then
   echo "Input illegal number of parameters " $#
-  echo "Need 3 parameters for the GPUs, the architecture, and the dataset-name"
+  echo "Need 4 parameters for the GPUs, the architecture, and the dataset-name, and the cutout"
   exit 1               
 fi 
 if [ "$TORCH_HOME" = "" ]; then
@@ -15,13 +15,14 @@ fi
 gpus=$1
 arch=$2
 dataset=$3
-SAVED=./snapshots/NAS/${arch}-${dataset}-E600
+cutout=$4
+SAVED=./snapshots/NAS/${arch}-${dataset}-${cutout}-E600
 
-CUDA_VISIBLE_DEVICES=${gpus} python ./exps-nas/train_base.py \
+CUDA_VISIBLE_DEVICES=${gpus} python ./exps-cnn/train_base.py \
 	--data_path $TORCH_HOME/cifar.python \
 	--dataset ${dataset} --arch ${arch} \
 	--save_path ${SAVED} \
 	--grad_clip 5 \
 	--init_channels 36 --layers 20 \
-	--model_config ./configs/nas-cifar-cos.config \
+	--model_config ./configs/nas-cifar-cos-${cutout}.config \
 	--print_freq 100 --workers 8