File size: 8,716 Bytes
639f3bb |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 |
"""
Example script to test different model backends
Demonstrates how to configure and use various model types
"""
import os
import asyncio
import sys
import time
from pathlib import Path
# Add the app directory to the Python path
sys.path.insert(0, str(Path(__file__).parent.parent))
from app.core.config import Settings
from app.services.model_backends.local_hf import LocalHuggingFaceBackend
from app.services.model_backends.hf_api import HuggingFaceAPIBackend
from app.services.model_backends.openai_api import OpenAIAPIBackend
from app.services.model_backends.anthropic_api import AnthropicAPIBackend
from app.models.schemas import ChatMessage
async def test_local_hf_backend():
"""Test local HuggingFace backend"""
print("🤖 Testing Local HuggingFace Backend")
print("-" * 40)
# Use a small model for testing
model_name = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"
backend = LocalHuggingFaceBackend(
model_name=model_name,
device="cpu", # Use CPU for compatibility
temperature=0.7,
max_tokens=50
)
try:
print(f"Loading model: {model_name}")
success = await backend.load_model()
if not success:
print("❌ Failed to load model")
return False
print("✅ Model loaded successfully")
# Test generation
messages = [
ChatMessage(role="user", content="Hello! What's your name?")
]
print("Generating response...")
start_time = time.time()
response = await backend.generate_response(messages, max_tokens=30)
end_time = time.time()
print(f"✅ Response generated in {end_time - start_time:.2f}s")
print(f"Response: {response.message}")
# Test streaming
print("\nTesting streaming...")
full_response = ""
chunk_count = 0
async for chunk in backend.generate_stream(messages, max_tokens=30):
full_response += chunk.content
chunk_count += 1
if chunk.is_final:
break
print(f"✅ Streaming completed with {chunk_count} chunks")
print(f"Streamed response: {full_response}")
# Cleanup
await backend.unload_model()
print("✅ Model unloaded")
return True
except Exception as e:
print(f"❌ Local HF backend test failed: {e}")
return False
async def test_hf_api_backend():
"""Test HuggingFace API backend"""
print("\n🌐 Testing HuggingFace API Backend")
print("-" * 40)
# Check if API token is available
api_token = os.getenv("HF_API_TOKEN")
if not api_token:
print("⚠️ HF_API_TOKEN not set, skipping HF API test")
return True
model_name = "microsoft/DialoGPT-medium"
backend = HuggingFaceAPIBackend(
model_name=model_name,
api_token=api_token,
temperature=0.7,
max_tokens=50
)
try:
print(f"Initializing API client for: {model_name}")
success = await backend.load_model()
if not success:
print("❌ Failed to initialize API client")
return False
print("✅ API client initialized")
# Test generation
messages = [
ChatMessage(role="user", content="Hello! How are you?")
]
print("Generating response via API...")
start_time = time.time()
response = await backend.generate_response(messages, max_tokens=30)
end_time = time.time()
print(f"✅ Response generated in {end_time - start_time:.2f}s")
print(f"Response: {response.message}")
return True
except Exception as e:
print(f"❌ HF API backend test failed: {e}")
return False
async def test_openai_backend():
"""Test OpenAI API backend"""
print("\n🔥 Testing OpenAI API Backend")
print("-" * 40)
# Check if API key is available
api_key = os.getenv("OPENAI_API_KEY")
if not api_key:
print("⚠️ OPENAI_API_KEY not set, skipping OpenAI test")
return True
model_name = "gpt-3.5-turbo"
backend = OpenAIAPIBackend(
model_name=model_name,
api_key=api_key,
temperature=0.7,
max_tokens=50
)
try:
print(f"Initializing OpenAI client for: {model_name}")
success = await backend.load_model()
if not success:
print("❌ Failed to initialize OpenAI client")
return False
print("✅ OpenAI client initialized")
# Test generation
messages = [
ChatMessage(role="user", content="Hello! What's the weather like?")
]
print("Generating response via OpenAI...")
start_time = time.time()
response = await backend.generate_response(messages, max_tokens=30)
end_time = time.time()
print(f"✅ Response generated in {end_time - start_time:.2f}s")
print(f"Response: {response.message}")
# Test streaming
print("\nTesting streaming...")
full_response = ""
chunk_count = 0
async for chunk in backend.generate_stream(messages, max_tokens=30):
full_response += chunk.content
chunk_count += 1
if chunk.is_final:
break
print(f"✅ Streaming completed with {chunk_count} chunks")
print(f"Streamed response: {full_response}")
return True
except Exception as e:
print(f"❌ OpenAI backend test failed: {e}")
return False
async def test_anthropic_backend():
"""Test Anthropic API backend"""
print("\n🧠 Testing Anthropic API Backend")
print("-" * 40)
# Check if API key is available
api_key = os.getenv("ANTHROPIC_API_KEY")
if not api_key:
print("⚠️ ANTHROPIC_API_KEY not set, skipping Anthropic test")
return True
model_name = "claude-3-haiku-20240307"
backend = AnthropicAPIBackend(
model_name=model_name,
api_key=api_key,
temperature=0.7,
max_tokens=50
)
try:
print(f"Initializing Anthropic client for: {model_name}")
success = await backend.load_model()
if not success:
print("❌ Failed to initialize Anthropic client")
return False
print("✅ Anthropic client initialized")
# Test generation
messages = [
ChatMessage(role="user", content="Hello! Tell me about yourself.")
]
print("Generating response via Anthropic...")
start_time = time.time()
response = await backend.generate_response(messages, max_tokens=30)
end_time = time.time()
print(f"✅ Response generated in {end_time - start_time:.2f}s")
print(f"Response: {response.message}")
return True
except Exception as e:
print(f"❌ Anthropic backend test failed: {e}")
return False
async def main():
"""Main test function"""
print("🚀 Sema Chat Backend Testing")
print("=" * 50)
results = {}
# Test each backend
results["local_hf"] = await test_local_hf_backend()
results["hf_api"] = await test_hf_api_backend()
results["openai"] = await test_openai_backend()
results["anthropic"] = await test_anthropic_backend()
# Summary
print("\n" + "=" * 50)
print("📊 Test Results Summary")
print("-" * 25)
for backend, success in results.items():
status = "✅ PASS" if success else "❌ FAIL"
print(f"{backend:15} {status}")
total_tests = len(results)
passed_tests = sum(results.values())
print(f"\nTotal: {passed_tests}/{total_tests} backends working")
if passed_tests == total_tests:
print("🎉 All available backends are working!")
elif passed_tests > 0:
print("⚠️ Some backends are working, check configuration for others")
else:
print("❌ No backends are working, check your setup")
print("\n💡 Tips:")
print("- For HF API: Set HF_API_TOKEN environment variable")
print("- For OpenAI: Set OPENAI_API_KEY environment variable")
print("- For Anthropic: Set ANTHROPIC_API_KEY environment variable")
print("- For local models: Ensure you have enough RAM/VRAM")
if __name__ == "__main__":
asyncio.run(main())
|