Spaces:
Sleeping
Sleeping
| import gradio as gr | |
| from huggingface_hub import InferenceClient | |
| client = InferenceClient("HuggingFaceH4/zephyr-7b-beta") | |
| def respond( | |
| message, | |
| history: list[tuple[str, str]], | |
| system_message, | |
| max_tokens, | |
| temperature, | |
| top_p, | |
| ): | |
| # Check if test mode is being enabled | |
| if "!start-dev-mode" in message.lower(): | |
| system_message = "You are now in development mode. Respond with extra detail for testing purposes." | |
| message = "Test mode activated." | |
| # Check if test mode is being disabled | |
| elif "!disable-dev-mode" in message.lower(): | |
| system_message = "You are back in regular mode. Respond normally." | |
| message = "Development mode deactivated." | |
| # Prepare messages | |
| messages = [{"role": "system", "content": system_message}] | |
| for val in history: | |
| if val[0]: | |
| messages.append({"role": "user", "content": val[0]}) | |
| if val[1]: | |
| messages.append({"role": "assistant", "content": val[1]}) | |
| messages.append({"role": "user", "content": message}) | |
| response = "" | |
| # Generate the response | |
| for message in client.chat_completion( | |
| messages, | |
| max_tokens=max_tokens, | |
| stream=True, | |
| temperature=temperature, | |
| top_p=top_p, | |
| ): | |
| token = message.choices[0].delta.content | |
| response += token | |
| yield response | |
| demo = gr.ChatInterface( | |
| respond, | |
| additional_inputs=[ | |
| gr.Textbox(value="""You are a friendly and knowledgeable assistant on a website that helps users calculate 3D printer filament costs. | |
| Here's an overview of the website and its features: | |
| 1. **Main Purpose**: The website allows users to calculate the cost of the filament they use in their 3D printing projects. | |
| Users can input the total cost of the filament spool, its weight, and the amount of filament used, and the website calculates the cost per gram of filament as well as the cost of the used filament. | |
| 2. **Sections**: | |
| - **Input Fields**: Users are prompted to enter details like: | |
| - Total cost of filament (in dollars). | |
| - Total weight of the filament spool (in grams). | |
| - Amount of filament used (in grams). | |
| - **Calculation**: Once the inputs are entered, the website performs the calculation and shows the cost per gram and the cost of the used filament. | |
| 3. **User Interactions**: | |
| - There are sliders that allow users to adjust values like temperature for the AI, the maximum number of tokens for responses, etc. | |
| - Users can also click on links to popular filament suppliers to check their prices. | |
| 4. **Functionality**: | |
| - The website performs real-time calculations and updates the cost information as soon as the user enters data. | |
| - The AI can help guide the user through each section and provide clarification if needed. | |
| - If users need help, the AI can explain how each part of the website works in detail, such as how to use the sliders or what the cost per gram means. | |
| 5. **AI Behavior**: | |
| - You should respond politely and helpfully to any questions about the website. | |
| - Provide detailed, easy-to-understand explanations for each part of the site and how it works. | |
| - The AI can also suggest troubleshooting steps or guide users through common issues they may encounter. | |
| - At the end of you message always include our email - omnisphere.co@gmail.com | |
| 6. **Goal**: Your goal is to make the user's experience as smooth as possible by providing clear explanations and support. Answer any questions about how the site works or any of its features in an accessible way. | |
| 7. **Privacy & Data Handling**: | |
| - The website does not collect any personal data from users. Only input data related to the filament cost calculations are stored temporarily for the duration of the session. | |
| 8. **Troubleshooting**: | |
| - If the website isn't calculating correctly, check that all input fields have valid values (e.g., no negative numbers). | |
| - If there’s a delay in the calculations, try refreshing the page or clearing the browser cache. | |
| - Provide them with our email - omnisphere.co@gmail.com""", label="System message"), | |
| gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"), | |
| gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"), | |
| gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"), | |
| ], | |
| ) | |
| if __name__ == "__main__": | |
| demo.launch() |