Update README.md

README.md

@@ -1,42 +1,27 @@
-# Mini Transformer Sentiment Model
-Uploaded via GitHub Actions
-
 ---
 language: en
-datasets:
-- custom
+license: mit
 tags:
 - sentiment-analysis
-license: mit
-library_name: pytorch
+- mini-transformer
 pipeline_tag: text-classification
+library_name: pytorch
+datasets:
+- custom
 ---
 
 # Mini Transformer Sentiment Model
 
-This is a lightweight Transformer encoder trained for sentiment analysis (positive/negative).  
-Built and uploaded automatically using a GitHub Actions pipeline.
-
-## 🧠 Model Details
-
-- Architecture: Mini Transformer Encoder
-- Framework: PyTorch
-- Task: Sentiment Classification
-- Input: Tokenized text
-- Output: Sentiment label (0 = negative, 1 = positive)
-
-## 🏋️‍♂️ Training
-
-Trained using a custom CSV dataset with sentences and binary sentiment labels.
+A minimal Transformer encoder for sentiment classification.  
+Trained on a small custom dataset using PyTorch.
 
-## 🧩 Usage
+## Usage
 
 ```python
 from transformers import AutoTokenizer, AutoModelForSequenceClassification
 
-model = AutoModelForSequenceClassification.from_pretrained("YOUR_USERNAME/mini-transformer-sentiment")
 tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
+model = AutoModelForSequenceClassification.from_pretrained("mishrabp/mini-transformers")
 
-inputs = tokenizer("I love this product!", return_tensors="pt")
-outputs = model(**inputs)
-print(outputs.logits)
+inputs = tokenizer("I love this!", return_tensors="pt")
+print(model(**inputs).logits)