Update app.py

app.py

@@ -16,78 +16,16 @@ tokenizer = AutoTokenizer.from_pretrained(model_path)
 def tokenize_text(input_text):
     return tokenizer.encode(input_text, truncation=False)
 
-# Function to summarize text in batches
-def process_batches(input_text):
-    tokens = tokenize_text(input_text)
-    text_tokens = len(tokens)
-    if text_tokens < 0.05 * TOKEN_LIMIT:
-        return "Error: Text too small to summarize."
-
-    batches = math.ceil(text_tokens / TOKEN_LIMIT)
-    summaries = []
-    
-    while batches > 1:
-        sentences = sent_tokenize(input_text)
-        avg_batch_size = math.ceil(len(sentences) / batches)
-        text_batches = [
-            " ".join(sentences[i:i+avg_batch_size])
-            for i in range(0, len(sentences), avg_batch_size)
-        ]
-
-        batch_summaries = []
-        for batch in text_batches:
-            max_length = int((TOKEN_LIMIT / batches) * 0.9)
-            min_length = int((TOKEN_LIMIT / batches) * 0.5)
-            summary_output = text_summary(
-                batch,
-                min_length=min_length,
-                max_length=max_length
-            )
-            batch_summaries.append(summary_output[0]['summary_text'])
-        
-        input_text = " ".join(batch_summaries)
-        tokens = tokenize_text(input_text)
-        text_tokens = len(tokens)
-        batches = math.ceil(text_tokens / TOKEN_LIMIT)
-
-    if text_tokens < 0.05 * TOKEN_LIMIT:
-        return "Error: Text too small to summarize."
-    
-    return input_text
-
-# Gradio button to set max/min lengths
-user_lengths = {"min_length": None, "max_length": None}
-
-def set_max_min_lengths(latest_text):
-    tokens = tokenize_text(latest_text)
-    text_tokens = len(tokens)
-    max_range = int(0.8 * text_tokens)
-    min_range = int(0.2 * text_tokens)
-    return f"Set max_length between {min_range} and {max_range} tokens."
-
-def validate_lengths(min_length, max_length, text):
-    if not (isinstance(min_length, int) and isinstance(max_length, int)):
-        return "Error: Length values must be integers."
-    if min_length <= 0 or max_length <= 0 or min_length >= max_length:
-        return "Error: Invalid length values. Ensure min_length < max_length and both are positive."
-    user_lengths["min_length"] = min_length
-    user_lengths["max_length"] = max_length
-    return "Done"
-
-def summarize_text(input_text):
-    if user_lengths["min_length"] is None or user_lengths["max_length"] is None:
-        return "Error: Please set valid max and min lengths first."
+# Function to summarize text
+def summarize_text(input_text, min_length=50, max_length=200):
     summary_output = text_summary(
         input_text,
-        min_length=user_lengths["min_length"],
-        max_length=user_lengths["max_length"]
+        min_length=min_length,
+        max_length=max_length
     )
     return summary_output[0]['summary_text']
 
-# Close any existing Gradio interfaces
-gr.close_all()
-
-# Article Example
+# Preload article and summary
 article_example = """Niteesh Nigam: A Visionary in Robotics, Machine Vision, and AI
 
 Niteesh Nigam, a forward-thinking robotics engineer and AI developer, has consistently demonstrated his passion for innovation and his ability to transform complex technological concepts into impactful real-world solutions. With a Master’s degree in Robotics and Autonomous Systems from Arizona State University (ASU) and a Bachelor’s degree in Mechanical Engineering from the Birla Institute of Technology and Science, Pilani Dubai, Niteesh has cultivated a robust foundation in engineering, computer vision, and artificial intelligence. His interdisciplinary expertise and hands-on experience have made him a standout professional in fields such as robotics, machine vision, deep learning, and control systems.
@@ -135,33 +73,39 @@ Niteesh Nigam’s work embodies a perfect blend of technical mastery, innovation
 As he looks to the future, Niteesh remains committed to making meaningful contributions in robotics, AI, and automation, with a focus on scalable solutions that benefit industries and communities alike. His journey is a testament to the transformative potential of technology when guided by a visionary like him.
 """
 
+# Precomputed summary
+summary_example = summarize_text(article_example, min_length=50, max_length=150)
+
 # Create Gradio interface
 with gr.Blocks() as demo:
-    text_input = gr.Textbox(label="Input Text", lines=10, value=article_example)
-    analyze_button = gr.Button("Analyze Text")
-    analysis_output = gr.Textbox(label="Analysis Result")
-
-    max_length_input = gr.Number(label="Max Length")
-    min_length_input = gr.Number(label="Min Length")
-    set_lengths_button = gr.Button("Set Max and Min Lengths")
-    length_result = gr.Textbox(label="Length Validation Result")
+    gr.Markdown("## **Niteesh Nigam Portfolio: Summarization Showcase**")
+    gr.Markdown(
+        "This app showcases the ability to summarize complex content into concise information. Below is an example:"
+    )
+    text_input = gr.Textbox(
+        label="Input Text", 
+        lines=10, 
+        value=article_example, 
+        interactive=False
+    )
+    summary_output = gr.Textbox(
+        label="Summarized Output", 
+        lines=10, 
+        value=summary_example, 
+        interactive=False
+    )
 
+    gr.Markdown("## Try It Yourself!")
+    custom_input = gr.Textbox(label="Input Your Text", lines=10, placeholder="Paste your text here...")
     summarize_button = gr.Button("Summarize")
-    summary_output = gr.Textbox(label="Summary Output", lines=10)
-
-    def analyze_and_set_text(input_text):
-        latest_text = process_batches(input_text)
-        if latest_text.startswith("Error"):
-            return latest_text, ""
-        return latest_text, set_max_min_lengths(latest_text)
-
-    analyze_button.click(analyze_and_set_text, inputs=text_input, outputs=[analysis_output, length_result])
-    set_lengths_button.click(
-        validate_lengths,
-        inputs=[min_length_input, max_length_input, analysis_output],
-        outputs=length_result
+    custom_summary_output = gr.Textbox(label="Summary Output", lines=10)
+
+    # Button action for custom input
+    summarize_button.click(
+        summarize_text, 
+        inputs=[custom_input], 
+        outputs=[custom_summary_output]
     )
-    summarize_button.click(summarize_text, inputs=analysis_output, outputs=summary_output)
 
 # Launch the Gradio app
-demo.launch(share=False)
+demo.launch(share=False)