Update README.md

README.md

@@ -12,10 +12,10 @@ In biology, "targeting peptides" typically refer to "targeting signal peptides"
 **TarPepSubLoc-ESM2**  (TarPepSubLoc, Targeting Peptide Subcellular Localization) is a protein language model fine-tuned from [**ESM2**](https://github.com/facebookresearch/esm) pretrained model [(***facebook/esm2_t36_3B_UR50D***)](https://huggingface.co/facebook/esm2_t36_3B_UR50D) on a trageting peptides subcelluar localization dataset with five classes.   
 
 **TarPepSubLoc-ESM2** achieved the following results:  
-Train Loss: 0.0012  
-Train Accuracy: 0.9972  
-Validation Loss: 0.0319  
-Validation Accuracy: 0.9826  
+Train Loss: 0.0385  
+Train Accuracy: 0.9881  
+Validation Loss: 0.0566  
+Validation Accuracy: 0.9812  
 Epoch: 20 
 # The dataset for training **TarPepSubLoc-ESM2**
 The full dataset contains 13,005 protein sequences, including SP (2,697), MT (499), CH (227), TH (45), and Other (9,537).