Upload 2 files

app.py

@@ -0,0 +1,47 @@
+import gradio as gr
+import torch
+from peft import PeftModel
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import numpy as np
+
+base_model = AutoModelForSequenceClassification.from_pretrained("bert-base-uncased")
+model = PeftModel.from_pretrained(base_model, "katsuchi/bert-base-uncased-twitter-sentiment-analysis")
+
+tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
+
+def get_sentiment(input_sentence):
+    inputs = tokenizer(input_sentence, return_tensors="pt", padding=True, truncation=True, max_length=512)
+    
+    inputs = {k: v.to(model.device) for k, v in inputs.items()}
+    
+    with torch.no_grad():
+        outputs = model(**inputs)
+        logits = outputs.logits
+    
+    probabilities = torch.nn.functional.softmax(logits, dim=-1).squeeze().cpu().numpy()
+    
+    labels = ["Negative", "Positive"]
+    
+    result = {labels[i]: round(prob, 3) for i, prob in enumerate(probabilities)}
+    
+    return result
+
+# Example sentences
+examples = [
+    ["I love this product!"],
+    ["This is the worst experience ever."],
+    ["The movie was okay, not great but not bad."],
+    ["Absolutely terrible, do not buy!"],
+    ["I feel amazing today!"]
+]
+
+iface = gr.Interface(
+    fn=get_sentiment,
+    inputs=gr.Textbox(label="Enter a sentence for sentiment analysis"),
+    outputs=gr.JSON(label="Sentiment Probabilities"),
+    title="Sentiment Analysis with Bert",
+    description="Enter a sentence, and this model will predict the sentiment (positive/negative) along with the probabilities.",
+    examples=examples
+)
+
+iface.launch()

requirements.txt

@@ -0,0 +1,5 @@
+gradio
+torch
+transformers
+peft
+numpy