cream codes update (#3383)

examples/nas/cream/lib/utils/builder_util.py

+import re
 import math
 import torch.nn as nn
 
+from copy import deepcopy
 from timm.utils import *
 from timm.models.layers.activations import Swish
 from timm.models.layers import CondConv2d, get_condconv_initializer

examples/nas/cream/lib/utils/util.py

@@ -4,6 +4,8 @@
 # email: haodu8-c@my.cityu.edu.hk and houwen.peng@microsoft.com
 
 import sys
+import logging
+import torch
 import argparse
 import torch.nn as nn
 

examples/nas/cream/test.py

@@ -83,7 +83,7 @@ def main():
                          'ir_r1_k5_s2_e4_c40_se0.25',
                          'ir_r1_k3_s2_e6_c80_se0.25',
                          'ir_r1_k3_s1_e6_c96_se0.25',
-                         'ir_r1_k3_s2_e6_c192_se0.25']
+                         'ir_r1_k5_s2_e6_c192_se0.25']
     arch_def = [[stem[0]]] + [[choice_block_pool[idx]
                                for repeat_times in range(len(arch_list[idx + 1]))]
                               for idx in range(len(choice_block_pool))] + [[stem[1]]]

nni/algorithms/nas/pytorch/cream/trainer.py

@@ -230,7 +230,7 @@ class CreamSupernetTrainer(Trainer):
         self.optimizer.zero_grad()
         grad_student_val = torch.autograd.grad(
             validation_loss,
-            self.model.module.rand_parameters(self.random_cand),
+            self.model.module.rand_parameters(self.current_student_arch),
             retain_graph=True)
 
         grad_teacher = torch.autograd.grad(
@@ -385,7 +385,7 @@ class CreamSupernetTrainer(Trainer):
                             step + 1, len(self.train_loader), meters)
 
         if self.main_proc and self.num_epochs == epoch + 1:
-            for idx, i in enumerate(self.best_children_pool):
+            for idx, i in enumerate(self.prioritized_board):
                 logger.info("No.%s %s", idx, i[:4])
 
     def validate_one_epoch(self, epoch):
@@ -396,9 +396,9 @@ class CreamSupernetTrainer(Trainer):
                 self.mutator.reset()
                 logits = self.model(x)
                 loss = self.val_loss(logits, y)
-                prec1, prec5 = self.accuracy(logits, y, topk=(1, 5))
+                prec1, prec5 = accuracy(logits, y, topk=(1, 5))
                 metrics = {"prec1": prec1, "prec5": prec5, "loss": loss}
-                metrics = self.reduce_metrics(metrics, self.distributed)
+                metrics = reduce_metrics(metrics)
                 meters.update(metrics)
 
                 if self.log_frequency is not None and step % self.log_frequency == 0:

nni/algorithms/nas/pytorch/cream/utils.py

@@ -17,7 +17,7 @@ def accuracy(output, target, topk=(1,)):
     if target.ndimension() > 1:
         target = target.max(1)[1]
 
-    correct = pred.eq(target.view(1, -1).expand_as(pred))
+    correct = pred.eq(target.reshape(1, -1).expand_as(pred))
 
     res = []
     for k in topk: