app.py

app.py

@@ -1,23 +1,40 @@
 import torch
 import gradio as gr
-# Use a pipeline as a high-level helper
 from transformers import pipeline
 
-# Initialize the summarization pipeline
-pipe = pipeline("summarization", model="Falconsai/text_summarization", torch_dtype=torch.float16)
+# Initialize the summarization pipeline (without float16 for CPU)
+pipe = pipeline("summarization", model="Falconsai/text_summarization")
+
+# Store the history
+history = []
 
 # Define the summarize function
-def summarize(input):
+def summarize(input, clear_history=False):
+    # Clear history if the flag is set
+    if clear_history:
+        history.clear()
+        return "History cleared!"
+
+    # Get the summary
     output = pipe(input)
-    return output[0]['summary_text']
+    summary = output[0]['summary_text']
+    
+    # Store the summary in history
+    history.append(summary)
+    
+    # Return the summary along with the history
+    return summary, history
 
 # Define the Gradio interface
 iface = gr.Interface(
     fn=summarize,
-    inputs=gr.Textbox(lines=10, placeholder="Enter text to summarize here..."),
-    outputs="text",
+    inputs=[
+        gr.Textbox(lines=10, placeholder="Enter text to summarize here..."),
+        gr.Checkbox(label="Clear history", value=False)  # Checkbox to clear history
+    ],
+    outputs=["text", "json"],  # Show summary and history in json format
     title="Text Summarizer",
-    description="Enter a long piece of text, and the summarizer will provide a concise summary."
+    description="Enter a long piece of text, and the summarizer will provide a concise summary. You can also clear the history of summaries."
 )
 
 # Launch the interface