NetPulse AI
Developed by Team HELIX AI
This project monitors network health using a Raspberry Pi, collecting data on CPU usage, temperature, signal strength, and packet loss. The data is logged in Excel, identifying abnormal conditions. Users upload the data to a Hugging Face chatbot for predictive analysis and optimization recommendations.
Features:
- Future Performance Prediction: Identifies upcoming failure risks based on past data trends.
- Risk Analysis and Potential Issues: Detects high-risk periods (e.g., peak CPU load at specific hours). Identifies network bottlenecks, low signal strength, or overheating components.
- Preventive Actions and Recommendations: Suggests cooling measures if temperature spikes are detected. Recommends bandwidth optimization if packet loss is increasing. Alerts users about critical failures and suggests preventive maintenance. Libraries Used:
- Gradio: For creating the user interface.
- Pandas: For reading and analyzing Excel files.
- Hugging API and LLM: Zephyr-7b-beta For utilizing state-of-the-art language models.
How It Works:
- Detailed Analysis
- Upload your data in form of excel file.
- Paste it in the Detailed Analysis Tab.
- Get detailed recommendations!
- General Chat for Network Optimization
- Talk to AI for more suggestions and queries regarding Network Issues.
Detailed Analysis
Analyze network performance trends, predict potential issues, and receive tailored recommendations for optimization based on the uploaded data.
1 2048
0.1 4
0.1 1
Upload Data
General Chat for Network Optimization
A chatbot that provides predictive maintenance insights, cost optimization suggestions, and energy efficiency recommendations.
1 2048
0.1 4
0.1 1