Upload app.py

app.py

@@ -0,0 +1,110 @@
+# -*- coding: utf-8 -*-
+"""Untitled6.ipynb
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1F6f_vJbssO7C2FM6FILWljFYacDmbVBY
+"""
+
+from IPython.display import display, HTML
+
+# Inject CSS to enable wrapping
+display(HTML('''
+    <style>
+        .output_area pre {
+            white-space: pre-wrap;
+        }
+    </style>
+'''))
+
+!pip install transformers gradio
+
+import os
+from google.colab import userdata
+
+# Get the Hugging Face API key from Colab Secrets
+api_key = userdata.get('HF_TOKEN')
+
+# Ensure the API key is set before using it
+if api_key is None:
+    raise ValueError("Hugging Face API key not found. Please ensure it is set in Colab Secrets.")
+
+# Set the Hugging Face token as an environment variable
+os.environ["HF_TOKEN"] = api_key
+
+# Hugging Face libraries will automatically use this token for authentication
+print("Hugging Face API key successfully loaded! You're good to go!")
+
+# Now you can continue with your Hugging Face-related code
+
+# Import necessary libraries
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Load model and tokenizer
+model_name = "distilgpt2"  # A lightweight, CPU-friendly model
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name)
+
+# Define the function to generate a response
+def generate_response(prompt):
+    # Tokenize the input prompt
+    inputs = tokenizer(prompt, return_tensors="pt")
+    # Generate a response
+    outputs = model.generate(
+    inputs.input_ids,
+    max_length=50,
+    do_sample=True,  # Enable sampling
+    temperature=0.7,  # Controls randomness
+    top_p=0.9,        # Nucleus sampling
+    pad_token_id=tokenizer.eos_token_id
+)
+
+    # Decode the output and set clean_up_tokenization_spaces to True to avoid warnings
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True, clean_up_tokenization_spaces=True)
+    return response
+
+# Example usage
+prompt = "I went to Safeway and I bought a"
+response = generate_response(prompt)
+print(response)
+
+def persona_response(prompt, persona="I am a helpful assistant"):
+    full_prompt = f"{persona}. {prompt}"
+    return generate_response(full_prompt)
+
+# Import Gradio
+import gradio as gr
+
+# Define Gradio interface function
+def chat_interface(user_input, persona="I am a helpful assistant"):
+    return persona_response(user_input, persona)
+
+# Set up Gradio interface
+interface = gr.Interface(
+    fn=chat_interface,
+    inputs=["text", "text"],  # Allows input for both prompt and persona
+    outputs="text",
+    title="Simple Chatbot",
+    description="Type something to chat with the bot! Add a persona to change its style, like 'I am a shopping assistant.'"
+)
+
+# Launch the Gradio interface in Colab
+interface.launch(share=True)  # share=True creates a public link
+
+"""### Uploading to Hugging Face Spaces
+
+Now that we have our chatbot working, here’s how to upload it to Hugging Face Spaces:
+
+1. Go to [Hugging Face Spaces](https://huggingface.co/spaces).
+2. Create a new Space, choose "Gradio" as the app type, and name it (e.g., "My Simple Chatbot").
+3. Upload this notebook, along with any necessary files or model assets.
+4. Set up your Space and click "Deploy." Your Gradio chatbot will now be live!
+
+**Tips for Deployment:**
+- Ensure you include any persona settings or customizations.
+- Test the app after deployment to confirm it works as expected.
+
+Once deployed, students can share the link with family and friends!
+
+"""