Update README.md

README.md

@@ -17,13 +17,11 @@ widget:
   example_title: "Positive Example"
 - text: "Terrible film, complete waste of time and money."
   example_title: "Negative Example"
-- text: "It was okay, nothing special but not bad either."
-  example_title: "Neutral Example"
 ---
 
 # DistilBERT Sentiment Analysis Model
 
-This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) for **3-class sentiment analysis** (Positive, Negative, Neutral) on movie reviews.
+This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) for **3-class sentiment analysis** (Positive, Negative) on movie reviews.
 
 ## 🎯 Model Description
 
@@ -31,7 +29,7 @@ This model is a fine-tuned version of [distilbert-base-uncased](https://huggingf
 - **Base Architecture:** DistilBERT (Distilled BERT)
 - **Language:** English
 - **Task:** Sentiment Analysis
-- **Classes:** 3 (Negative, Neutral, Positive)
+- **Classes:** 2 (Negative, Positive)
 - **Parameters:** ~66M
 - **Model Size:** ~250MB
 
@@ -87,7 +85,7 @@ with torch.no_grad():
 predicted_class = torch.argmax(predictions, dim=-1).item()
 confidence = predictions[0][predicted_class].item()
 
-labels = ["NEGATIVE", "NEUTRAL", "POSITIVE"]
+labels = ["NEGATIVE", POSITIVE"]
 print(f"Sentiment: {labels[predicted_class]} (Confidence: {confidence:.3f})")
 ```
 
@@ -126,9 +124,8 @@ print(f"Sentiment: {labels[predicted_class]} (Confidence: {confidence:.3f})")
 | Validation Loss | 0.18 |
 
 ### Class Distribution
-- **Negative:** 33.3% (1,667 samples)
-- **Neutral:** 33.3% (1,667 samples) 
-- **Positive:** 33.3% (1,666 samples)
+- **Negative:** 33.3% (2500 samples)
+- **Positive:** 33.3% (2500 samples)
 
 ## 🎯 Intended Use
 
@@ -151,7 +148,6 @@ print(f"Sentiment: {labels[predicted_class]} (Confidence: {confidence:.3f})")
 - **Domain Specificity:** Primarily trained on movie reviews, may not generalize well to other domains
 - **Language:** English only, no multilingual support
 - **Context Length:** Limited to 256 tokens, longer texts are truncated
-- **Neutral Class:** Synthetic neutral samples may not represent real-world neutral sentiment accurately
 - **Cultural Bias:** May reflect biases present in IMDB dataset
 
 ### Potential Biases
@@ -188,7 +184,7 @@ DistilBERT Base
 ### Output Format
 ```python
 {
-    'label': 'POSITIVE',  # One of: NEGATIVE, NEUTRAL, POSITIVE
+    'label': 'POSITIVE',  # One of: NEGATIVE, POSITIVE
     'score': 0.9987       # Confidence score (0-1)
 }
 ```