correlation.py

@@ -35,8 +35,8 @@ parser = argparse.ArgumentParser()
 
 # general setting
 parser.add_argument('--data_path', default="datasets", type=str, nargs='?', help='path to the image dataset (datasets or datasets/ILSVRC/Data/CLS-LOC)')
-parser.add_argument('--seed', default=0, type=int, help='random seed')
-parser.add_argument('--device', default="cuda", type=str, nargs='?', help='setup device (cpu, mps or cuda)')
+parser.add_argument('--seed', default=111, type=int, help='random seed')
+parser.add_argument('--device', default="cuda:1", type=str, nargs='?', help='setup device (cpu, mps or cuda)')
 parser.add_argument('--repeats', default=32, type=int, nargs='?', help='times of calculating the training-free metric')
 parser.add_argument('--input_samples', default=16, type=int, nargs='?', help='input batch size for training-free metric')
 
@@ -93,14 +93,11 @@ if __name__ == "__main__":
         print(f'Average SWAP score: {np.mean(swap_score)}')
         print(f'Elapsed time: {end_time - start_time:.2f} seconds')
 
-        results.append([np.mean(swap_score), acc, i])
-
-    results = pd.DataFrame(results, columns=['swap_score', 'valid_acc', 'index'])
+        results.append([np.mean(swap_score), acc])
 
+    results = pd.DataFrame(results, columns=['swap_score', 'valid_acc'])
     print()    
     print(f'Spearman\'s Correlation Coefficient: {stats.spearmanr(results.swap_score, results.valid_acc)[0]}')
-    results.to_csv('swap_results.csv', float_format='%.4f', index=False)
-