Create app.py

app.py

@@ -0,0 +1,58 @@
+import gradio as gr
+import torch
+import pandas as pd
+from transformers import T5ForConditionalGeneration, T5Tokenizer
+import io
+
+# Load the pre-trained T5 model for paraphrasing
+model = T5ForConditionalGeneration.from_pretrained('ramsrigouthamg/t5_paraphraser')
+tokenizer = T5Tokenizer.from_pretrained('t5-base')
+
+def generate_paraphrases(text, num_return_sequences=5, num_beams=10):
+    input_text = "paraphrase: " + text + " </s>"
+    encoding = tokenizer.encode_plus(input_text, return_tensors="pt")
+    input_ids = encoding["input_ids"]
+
+    outputs = model.generate(
+        input_ids,
+        max_length=256,
+        num_beams=num_beams,
+        num_return_sequences=num_return_sequences,
+        no_repeat_ngram_size=2,
+        early_stopping=True
+    )
+
+    paraphrases = [
+        tokenizer.decode(output, skip_special_tokens=True, clean_up_tokenization_spaces=True)
+        for output in outputs
+    ]
+    return paraphrases
+
+def generate_and_export(text):
+    paraphrases = generate_paraphrases(text)
+    df = pd.DataFrame({
+        "Original": [text] * len(paraphrases),
+        "Perspective": paraphrases
+    })
+    csv_buffer = io.StringIO()
+    df.to_csv(csv_buffer, index=False)
+    csv_bytes = csv_buffer.getvalue().encode()
+    return paraphrases, csv_bytes
+
+def gradio_generate(text):
+    paraphrases, csv_bytes = generate_and_export(text)
+    output_text = "\n\n".join(f"Perspective {i+1}: {p}" for i, p in enumerate(paraphrases))
+    return output_text, (csv_bytes, "generated_perspectives.csv")
+
+iface = gr.Interface(
+    fn=gradio_generate,
+    inputs=gr.Textbox(lines=5, placeholder="Enter text here..."),
+    outputs=[
+        gr.Textbox(label="Generated Perspectives"),
+        gr.File(label="Download CSV")
+    ],
+    title="Paraphrase Perspective Generator",
+    description="Enter any text and generate alternative perspectives (paraphrases). Download the results as a CSV file."
+)
+
+iface.launch()