Update README.md

README.md

@@ -5,7 +5,7 @@ The following code load and test the models on colab notebook.
 
 ---
 
-## Prerequisites
+# Step 1: Prerequisites
 
 1. Import the required Python packages:
 
@@ -28,10 +28,11 @@ from huggingface_hub import login
 login("Replace with the key")
 ```
 
-# Define the preprocessing and dataset class
+# Step 2: Define the preprocessing and dataset class
 
-1. Run the following preprocessing code 
+1. Run the following class and functions designed to preprocess the test data
 
+```python
 class NewsDataset(Dataset):
     def __init__(self, texts, labels, tokenizer, max_len=128):
         self.texts = texts
@@ -75,33 +76,41 @@ def preprocess_text(text):
     text = text.lower()
     text = ' '.join(text.split())
     return text
+```
 
 
-
-# Step 1: Load the model and tokenizer from Hugging Face Hub
+# Step 3: Load the model and tokenizer from Hugging Face Hub
+```python
 print("Loading model and tokenizer...")
-REPO_NAME = "CIS5190GoGo/CustomModel"  # Replace with your repo name on Hugging Face Hub
+REPO_NAME = "CIS5190GoGo/CustomModel" #This is where we pushed the model to
 model = RobertaForSequenceClassification.from_pretrained(REPO_NAME)
 tokenizer = RobertaTokenizer.from_pretrained(REPO_NAME)
 
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 model.to(device)
 print("Model and tokenizer loaded successfully!")
+```
 
-# Step 2: Load test dataset
+# Step 4: Load test dataset
+```python
 print("Loading test data...")
 test_data_path = "/content/drive/MyDrive/5190_project/test_data_random_subset.csv"  # Replace with your test set path
 test_data = pd.read_csv(test_data_path)
-
-# Preprocess test data
+```
+# Step 5: Preprocess test data
+```python
 X_test = test_data['title'].apply(preprocess_text).values
 y_test = test_data['labels'].values
+```
 
-# Step 3: Prepare the dataset and dataloader
+# Step 6: Prepare the dataset and dataloader
+```python
 test_dataset = NewsDataset(X_test, y_test, tokenizer)
 test_loader = DataLoader(test_dataset, batch_size=16, num_workers=2)
+```
 
-# Step 4: Evaluate the model
+# Step 7: Evaluate the model and calculate accuracy
+```python
 print("Evaluating the model...")
 model.eval()
 all_preds, all_labels = [], []
@@ -118,6 +127,6 @@ with torch.no_grad():
         all_preds.extend(preds.cpu().numpy())
         all_labels.extend(labels.cpu().numpy())
 
-# Step 5: Calculate accuracy
 accuracy = accuracy_score(all_labels, all_preds)
-print(f"Test Accuracy: {accuracy:.4f}")
+print(f"Test Accuracy: {accuracy:.4f}")
+```