Update README.md

README.md

@@ -64,54 +64,67 @@ Epoch 1/10
 train Loss: 1.0083 Acc: 0.6850
 valid Loss: 0.6304 Acc: 0.7985
 Epoch 1 completed in 2109.20 seconds.
+
 Epoch 2/10
 ----------
 train Loss: 0.7347 Acc: 0.7687
 valid Loss: 0.8616 Acc: 0.7307
 Epoch 2 completed in 2183.41 seconds.
+
 Epoch 3/10
 ----------
 train Loss: 0.6510 Acc: 0.7913
 valid Loss: 0.5594 Acc: 0.8260
 Epoch 3 completed in 2174.55 seconds.
+
 Epoch 4/10
 ----------
 train Loss: 0.5762 Acc: 0.8126
 valid Loss: 0.4006 Acc: 0.8655
 Epoch 4 completed in 2166.46 seconds.
+
 Epoch 5/10
 ----------
 train Loss: 0.5478 Acc: 0.8210
 valid Loss: 0.3968 Acc: 0.8793
 Epoch 5 completed in 2189.89 seconds.
+
 Epoch 6/10
 ----------
 train Loss: 0.5223 Acc: 0.8272
 valid Loss: 0.4051 Acc: 0.8729
 Epoch 6 completed in 2185.71 seconds.
+
 Epoch 7/10
 ----------
 train Loss: 0.4974 Acc: 0.8355
 valid Loss: 0.3223 Acc: 0.9094
 Epoch 7 completed in 2184.83 seconds.
+
 Epoch 8/10
 ----------
 train Loss: 0.3464 Acc: 0.8870
 valid Loss: 0.2221 Acc: 0.9338
 Epoch 8 completed in 2184.53 seconds.
+
 Epoch 9/10
 ----------
 train Loss: 0.2896 Acc: 0.9049
 valid Loss: 0.2125 Acc: 0.9338
 Epoch 9 completed in 2181.82 seconds.
+
 Epoch 10/10
 ----------
-
 train Loss: 0.2604 Acc: 0.9136
+
 valid Loss: 0.2076 Acc: 0.9326
+
 Training complete in 362m 11s
+
 Best val Acc: 0.9338
+
 Training Time: Approximately 12 minutes on a single GPU for 10 epochs.
+
 The model showed high accuracy in predicting common categories such as plastic, paper, and metal, but struggled with classes like shoes and clothes, reflecting the challenges of web-scraped images for such categories.
 
 ## Conclusion