Upload inference_api_example.py with huggingface_hub

inference_api_example.py

@@ -0,0 +1,72 @@
+#!/usr/bin/env python3
+# Example script for using the Hugging Face Inference API with the D&D model
+# This approach will track download metrics on Hugging Face
+
+import requests
+import json
+import time
+
+# Configuration
+API_URL = "https://api-inference.huggingface.co/models/chendren/deepseek-dnd-lora"
+headers = {
+    "Content-Type": "application/json",
+    # Replace with your Hugging Face API token
+    "Authorization": "Bearer YOUR_API_TOKEN"
+}
+
+# Test prompts
+test_prompts = [
+    "Create a D&D character with the following details: Race: Half-Elf, Class: Bard, Background: Entertainer",
+    "Design a D&D adventure hook set in a dark forest",
+    "Create a magical item for D&D 5e that would be suitable for a level 5 rogue",
+    "Write a description for a fantasy tavern in a D&D setting"
+]
+
+def query(payload):
+    """
+    Send a query to the Hugging Face API
+    """
+    response = requests.post(API_URL, headers=headers, json=payload)
+    
+    if response.status_code == 503:
+        # Model is loading
+        print("Model is loading. Waiting...")
+        time.sleep(20)
+        return query(payload)
+    
+    return response.json()
+
+# Run tests
+for i, prompt in enumerate(test_prompts):
+    print(f"\n==== Test Prompt {i+1} ====")
+    print(prompt)
+    print("\n==== Response ====")
+    
+    # Add a system prompt to help guide the model
+    full_prompt = f"You are a Dungeons & Dragons assistant. {prompt}"
+    
+    # Make the API request - this will be tracked by HF Hub
+    payload = {
+        "inputs": full_prompt,
+        "parameters": {
+            "max_new_tokens": 500,
+            "temperature": 0.7,
+            "top_p": 0.9,
+            "top_k": 50,
+            "repetition_penalty": 1.1,
+            "do_sample": True
+        }
+    }
+    
+    try:
+        result = query(payload)
+        print(json.dumps(result, indent=2))
+    except Exception as e:
+        print(f"Error: {e}")
+    
+    print("\n" + "="*50)
+    
+    # Wait a bit between requests to avoid rate limiting
+    time.sleep(3)
+
+print("\nTesting complete!")