README.md

metadata

model-index:
  - name: Cloud Service RAG Model
    results:
      - task:
          type: text-generation
        dataset:
          name: rag_contents
          type: pricing_info
        metrics:
          - type: custom
            value: 85
            name: Retrieval Accuracy
      - task:
          type: retrieval-augmented generation (RAG)
        dataset:
          name: pricing_info
          type: custom
        metrics:
          - type: custom
            value: 75
            name: Pricing Retrieval Accuracy

☁️ Cloud Cents RAG Model ☁️

📖 Model Description

This model is a fine-tuned GPT-2 model designed to answer cloud-related questions about AWS, Azure, GCP, and other cloud platforms. It uses Retrieval-Augmented Generation (RAG) to combine document retrieval with text generation, leveraging cloud-related documents and real-time pricing information.

🛠 Intended Use

• 🔍 Cloud service comparisons (e.g., AWS vs Azure)
• 💰 Real-time cloud pricing queries
• 🌐 General information about cloud platforms and services

🔧 Pipeline

text-generation with retrieval-augmented generation (RAG)

• 📚 Document retrieval from rag_contents and pricing_info
• ✍️ Text generation using fine-tuned GPT-2

📊 Datasets

• rag_contents: Contains cloud-related documents from sources such as AWS, Azure, GCP.
• pricing_info: Provides real-time pricing details for cloud services (e.g., EC2, Blob Storage, Container Registry).

📈 Metrics

• Perplexity: Evaluated for fluency of text generation.
• BLEU: Used for measuring the accuracy of generated answers.
• Retrieval Accuracy: Custom metric for FAISS-based document retrieval.

⚠️ Limitations

• 🕑 The model may not always retrieve the most up-to-date information about cloud services.
• 📂 The retrieval and generation quality is based on the documents stored in the rag_contents table.