Getting ready to run FEAR on DARTS space without any freeze.

.vscode/launch.json

@@ -921,6 +921,15 @@
             "args": ["--nb301-logs-dir", "C:\\Users\\dedey\\dataroot\\nasbench301\\nasbench301_full_data\\nb_301_v13_lc_iclr_final\\rs",
                 "--out-dir", "F:\\archai_experiment_reports"]
         },
+        {
+            "name": "Analysis Simulate FEAR on Nasbench301",
+            "type": "python",
+            "request": "launch",
+            "program": "${cwd}/scripts/reports/fear_analysis/simulate_fear_on_nb301.py",
+            "console": "integratedTerminal",
+            "args": ["--nb301-logs-dir", "C:\\Users\\dedey\\dataroot\\nasbench301\\nasbench301_full_data\\nb_301_v13_lc_iclr_final\\rs",
+                "--out-dir", "F:\\archai_experiment_reports"]
+        },
         {
             "name": "CurrentFile",
             "type": "python",

archai/algos/proxynas/freeze_trainer.py

@@ -39,8 +39,18 @@ class FreezeTrainer(ArchTrainer, EnforceOverrides):
 
         # freeze everything other than the last layer
         if not self.conf_train['bypass_freeze']:
-            logger.info('no freezing!')
+            # addup parameters which are not frozen
+            num_frozen_params = 0
+            for l in model_stats.layer_stats:
+                for identifier in self.conf_train['identifiers_to_unfreeze']:
+                    if identifier in l.name:
+                        num_frozen_params += l.parameters            
+            ratio_unfrozen = num_frozen_params / model_stats.parameters 
+            logger.info(f'unfrozen parameters ratio {ratio_unfrozen}')
+
             self._freeze_but_last_layer()
+        else:
+            logger.info(f'Bypassing freezing!')
 
 
     def _freeze_but_last_layer(self) -> None:

confs/algos/random_dartsspace_far.yaml

@@ -90,7 +90,8 @@ nas:
       cv_num: 5 # total number of folds available
       dataset:
         _copy: '/dataset'
-    freeze_trainer:  
+    freeze_trainer:
+      bypass_freeze: True  
       identifiers_to_unfreeze: ['logits_op._op', 'cells.7', 'cells.6'] # last few layer names in DARTS space are 'logits_op._op': Linear, 'cells.19': prefix for all cell 19 parameters
       apex:
         _copy: '/common/apex'
@@ -125,7 +126,7 @@ nas:
         lossfn:
           type: 'CrossEntropyLoss'
     trainer_full:
-      top1_acc_threshold: 0.6
+      top1_acc_threshold: 0.1
       use_val: False
       apex:
         _copy: '/common/apex'

scripts/reports/fear_analysis/simulate_fear_on_nb301.py

@@ -14,7 +14,7 @@ import statistics
 def find_train_thresh_epochs(train_acc:List[float], train_thresh:float)->int:
     for i, t in enumerate(train_acc):
         if t >= train_thresh:
-            return i 
+            return i + 1
 
 
 def main():
@@ -24,13 +24,14 @@ def main():
     args, extra_args = parser.parse_known_args()
 
     train_thresh = 60.0
-    post_thresh_epochs = 5
+    post_thresh_epochs = 10
 
     all_test_acc = []
     all_fear_end_acc = []
     all_fear_time = []
 
     all_reg_train_acc = defaultdict(list)
+    all_reg_train_time_per_epoch = defaultdict(list)
     
     # collect all the json file names in the log dir recursively
     for root, dir, files in os.walk(args.nb301_logs_dir):
@@ -61,6 +62,7 @@ def main():
                 # evaluation baseline
                 for epoch_num, train_acc in enumerate(log_data['learning_curves']['Train/train_accuracy']):
                     all_reg_train_acc[epoch_num].append(train_acc)
+                    all_reg_train_time_per_epoch[epoch_num].append((epoch_num + 1) * per_epoch_time)                
 
 
 
@@ -69,14 +71,17 @@ def main():
     print(f'FEAR avg time: {statistics.mean(all_fear_time)}')
 
     spes_train_acc_vs_epoch = {}
+    avg_time_train_acc_vs_epoch = {}
     for epoch_num, train_accs_epoch in all_reg_train_acc.items():
         if len(train_accs_epoch) != len(all_test_acc):
             continue
         this_spe, _ = spearmanr(all_test_acc, train_accs_epoch)
         spes_train_acc_vs_epoch[epoch_num] = this_spe
+        avg_time_train_acc_vs_epoch[epoch_num] = statistics.mean(all_reg_train_time_per_epoch[epoch_num])
 
     for epoch_num, spe in spes_train_acc_vs_epoch.items():
-        print(f'Epoch {epoch_num}, spearman {spe}')                
+        avg_time = avg_time_train_acc_vs_epoch[epoch_num]
+        print(f'Epoch {epoch_num}, spearman {spe}, avg. time: {avg_time} seconds')                
 
 
 

scripts/reports/fear_plots/cross_exp_conf.yaml

@@ -4,8 +4,6 @@ exp_folder: 'F:\\archai_experiment_reports'
 darts_cifar10:
   freezetrain:
     ft_dt_fb96_ftlr0.025_fte10_ct96_ftt0.6_ftonly: 'ft_dt_fb96_ftlr0.025_fte10_ct96_ftt0.6_ftonly'
-<<<<<<< HEAD
-=======
     ft_dt_fb96_ftlr0.025_fte15_ct96_ftt0.6_ftonly: 'ft_dt_fb96_ftlr0.025_fte15_ct96_ftt0.6_ftonly'
     ft_dt_fb96_ftlr0.025_fte20_ct96_ftt0.6_ftonly: 'ft_dt_fb96_ftlr0.025_fte20_ct96_ftt0.6_ftonly'
     ft_dt_fb96_ftlr0.025_fte10_ct96_ftt0.6_c5_ftonly: 'ft_dt_fb96_ftlr0.025_fte10_ct96_ftt0.6_c5_ftonly'
@@ -13,7 +11,6 @@ darts_cifar10:
     ft_dt_fb96_ftlr0.025_fte10_ct96_ftt0.6_c3_ftonly: 'ft_dt_fb96_ftlr0.025_fte10_ct96_ftt0.6_c3_ftonly'
     ft_dt_fb96_ftlr0.025_fte10_ct96_ftt0.6_nofreeze_ftonly: 'ft_dt_fb96_ftlr0.025_fte10_ct96_ftt0.6_nofreeze_ftonly'
     
->>>>>>> eaaf94d5 (Added analysis script for ranking in nb301.)
 
   shortreg:
     dt_reg_b96_e5: 'dt_reg_b96_e5'