add epochs_no_optarch to replace method in darts/train_search (#215)

demo/darts/README.md

@@ -18,7 +18,7 @@
 ``` bash
 python search.py                       # DARTS一阶近似搜索方法
 python search.py --unrolled=True       # DARTS的二阶近似搜索方法
-python search.py --method='PC-DARTS'   # PC-DARTS搜索方法
+python search.py --method='PC-DARTS' --batch_size=256 --learning_rate=0.1 --arch_learning_rate=6e-4 --epochs_no_archopt=15   # PC-DARTS搜索方法
 ```
 
 模型结构随搜索轮数的变化如图1所示。需要注意的是，图中准确率Acc并不代表该结构最终准确率，为了获得当前结构的最佳准确率，请对得到的genotype做网络结构评估训练。
@@ -86,4 +86,4 @@ def train_search(batch_size, train_portion, is_shuffle, args):
 python visualize.py PC-DARTS
 ```
 
-`PC-DARTS`代表某个Genotype结构，需要预先添加到genotype.py中
\ No newline at end of file
+`PC-DARTS`代表某个Genotype结构，需要预先添加到genotype.py中

demo/darts/search.py

@@ -50,7 +50,8 @@ add_arg('trainset_num',      int,   50000,           "images number of trainset.
 add_arg('model_save_dir',    str,   'search_cifar', "The path to save model.")
 add_arg('grad_clip',         float, 5,               "Gradient clipping.")
 add_arg('arch_learning_rate',float, 3e-4,            "Learning rate for arch encoding.")
-add_arg('method',            str,   'DARTS',      "The search method you would like to use")
+add_arg('method',            str,   'DARTS',         "The search method you would like to use")
+add_arg('epochs_no_archopt', int,   0,               "Epochs not optimize the arch params")
 add_arg('cutout_length',     int,   16,              "Cutout length.")
 add_arg('cutout',            ast.literal_eval,  False, "Whether use cutout.")
 add_arg('unrolled',          ast.literal_eval,  False, "Use one-step unrolled validation loss")
@@ -84,8 +85,9 @@ def main(args):
             num_imgs=args.trainset_num,
             arch_learning_rate=args.arch_learning_rate,
             unrolled=args.unrolled,
-            method=args.method,
             num_epochs=args.epochs,
+            epochs_no_archopt=args.epochs_no_archopt,
+            use_gpu=args.use_gpu,
             use_data_parallel=args.use_data_parallel,
             log_freq=args.log_freq)
         searcher.train()

paddleslim/nas/darts/train_search.py

@@ -26,8 +26,6 @@ from ...common import AvgrageMeter, get_logger
 from .architect import Architect
 logger = get_logger(__name__, level=logging.INFO)
 
-SUPPORTED_METHODS = ["PC-DARTS", "DARTS"]
-
 
 def count_parameters_in_MB(all_params):
     parameters_number = 0
@@ -47,8 +45,8 @@ class DARTSearch(object):
                  num_imgs=50000,
                  arch_learning_rate=3e-4,
                  unrolled='False',
-                 method='DARTS',
                  num_epochs=50,
+                 epochs_no_archopt=0,
                  use_gpu=True,
                  use_data_parallel=False,
                  log_freq=50):
@@ -60,9 +58,7 @@ class DARTSearch(object):
         self.num_imgs = num_imgs
         self.arch_learning_rate = arch_learning_rate
         self.unrolled = unrolled
-        self.method = method
-        assert (self.method in SUPPORTED_METHODS
-                ), "Currently only support PC-DARTS, DARTS two methods"
+        self.epochs_no_archopt = epochs_no_archopt
         self.num_epochs = num_epochs
         self.use_gpu = use_gpu
         self.use_data_parallel = use_data_parallel
@@ -94,7 +90,7 @@ class DARTSearch(object):
             valid_label.stop_gradient = True
             n = train_image.shape[0]
 
-            if not (self.method == "PC-DARTS" and epoch < 15):
+            if epoch >= self.epochs_no_archopt:
                 architect.step(train_image, train_label, valid_image,
                                valid_label)