File size: 7,874 Bytes
e4d5155 |
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 |
"""
Benchmarking script for efficient-context performance.
"""
import logging
import time
import argparse
import random
import string
import psutil
import os
import gc
from typing import List, Dict, Any
from efficient_context import ContextManager
from efficient_context.compression import SemanticDeduplicator
from efficient_context.chunking import SemanticChunker
from efficient_context.retrieval import CPUOptimizedRetriever
# Set up logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
def generate_random_text(words: int = 1000, paragraphs: int = 5) -> str:
"""
Generate random text for benchmarking.
Args:
words: Number of words to generate
paragraphs: Number of paragraphs to split the text into
Returns:
text: Generated random text
"""
# List of common words for more realistic text
common_words = [
"the", "be", "to", "of", "and", "a", "in", "that", "have", "I",
"it", "for", "not", "on", "with", "he", "as", "you", "do", "at",
"this", "but", "his", "by", "from", "they", "we", "say", "her", "she",
"or", "an", "will", "my", "one", "all", "would", "there", "their", "what",
"so", "up", "out", "if", "about", "who", "get", "which", "go", "me",
"renewable", "energy", "climate", "wind", "solar", "power", "change", "global",
"sustainable", "resources", "efficiency", "emissions", "carbon", "technology"
]
# Generate paragraphs
result = []
words_per_paragraph = words // paragraphs
for i in range(paragraphs):
paragraph_words = []
for j in range(words_per_paragraph):
# Occasionally add a random word for variety
if random.random() < 0.1:
word = ''.join(random.choice(string.ascii_lowercase) for _ in range(random.randint(3, 10)))
else:
word = random.choice(common_words)
# Capitalize first word of sentence
if j == 0 or paragraph_words[-1].endswith('.'):
word = word.capitalize()
# Add punctuation occasionally
if j > 0 and j % random.randint(8, 15) == 0:
word += '.'
elif random.random() < 0.05:
word += ','
paragraph_words.append(word)
# Ensure paragraph ends with period
if not paragraph_words[-1].endswith('.'):
paragraph_words[-1] += '.'
result.append(' '.join(paragraph_words))
return '\n\n'.join(result)
def get_memory_usage() -> Dict[str, Any]:
"""
Get current memory usage.
Returns:
stats: Memory usage statistics
"""
process = psutil.Process(os.getpid())
memory_info = process.memory_info()
return {
"rss": memory_info.rss / (1024 * 1024), # MB
"vms": memory_info.vms / (1024 * 1024) # MB
}
def run_benchmark(
num_documents: int = 10,
words_per_document: int = 1000,
num_queries: int = 5
) -> None:
"""
Run a benchmark of efficient-context performance.
Args:
num_documents: Number of documents to process
words_per_document: Number of words per document
num_queries: Number of queries to run
"""
logger.info(f"Starting benchmark with {num_documents} documents, {words_per_document} words each")
# Initialize context manager
context_manager = ContextManager(
compressor=SemanticDeduplicator(threshold=0.85),
chunker=SemanticChunker(chunk_size=256),
retriever=CPUOptimizedRetriever(embedding_model="lightweight")
)
# Generate documents
logger.info("Generating random documents...")
documents = []
for i in range(num_documents):
content = generate_random_text(words=words_per_document, paragraphs=5)
documents.append({
"content": content,
"metadata": {"id": f"doc-{i}", "source": "benchmark"}
})
# Measure document processing
logger.info("Adding documents to context manager...")
start_mem = get_memory_usage()
start_time = time.time()
document_ids = context_manager.add_documents(documents)
end_time = time.time()
end_mem = get_memory_usage()
processing_time = end_time - start_time
memory_increase = end_mem["rss"] - start_mem["rss"]
logger.info(f"Document processing:")
logger.info(f" - Time: {processing_time:.2f} seconds")
logger.info(f" - Average per document: {processing_time / num_documents:.4f} seconds")
logger.info(f" - Memory usage increase: {memory_increase:.2f} MB")
logger.info(f" - Total chunks created: {len(context_manager.chunks)}")
# Generate random queries
logger.info("Generating context for queries...")
queries = [
f"Explain {random.choice(['renewable', 'sustainable', 'clean', 'alternative'])} energy",
f"What are the {random.choice(['benefits', 'advantages', 'impacts', 'effects'])} of {random.choice(['solar', 'wind', 'hydro', 'geothermal'])} power?",
f"How does {random.choice(['climate change', 'global warming', 'carbon emissions', 'greenhouse gases'])} affect the environment?",
f"Discuss the {random.choice(['future', 'potential', 'limitations', 'challenges'])} of renewable energy",
f"What is the {random.choice(['relationship', 'connection', 'link', 'correlation'])} between energy consumption and climate change?"
]
# Ensure we have enough queries
while len(queries) < num_queries:
queries.append(f"Tell me about {random.choice(['energy', 'climate', 'sustainability', 'emissions'])}")
# Select the requested number of queries
selected_queries = random.sample(queries, min(num_queries, len(queries)))
# Measure query processing
total_query_time = 0
total_query_tokens = 0
for i, query in enumerate(selected_queries):
# Clear some memory and cache before each query
gc.collect()
start_time = time.time()
context = context_manager.generate_context(query)
query_time = time.time() - start_time
context_tokens = len(context.split())
total_query_time += query_time
total_query_tokens += context_tokens
logger.info(f"Query {i+1}: '{query}'")
logger.info(f" - Time: {query_time:.4f} seconds")
logger.info(f" - Context size: {context_tokens} tokens")
avg_query_time = total_query_time / num_queries
avg_tokens = total_query_tokens / num_queries
logger.info("\nBenchmark Summary:")
logger.info(f" - Documents processed: {num_documents} ({words_per_document} words each)")
logger.info(f" - Queries executed: {num_queries}")
logger.info(f" - Document processing time: {processing_time:.2f} seconds ({processing_time / num_documents:.4f}s per document)")
logger.info(f" - Average query time: {avg_query_time:.4f} seconds")
logger.info(f" - Average context size: {avg_tokens:.1f} tokens")
logger.info(f" - Final memory usage: {get_memory_usage()['rss']:.2f} MB")
def main():
"""Main function for the benchmark script."""
parser = argparse.ArgumentParser(description="Benchmark efficient-context performance")
parser.add_argument("--documents", type=int, default=10, help="Number of documents to process")
parser.add_argument("--words", type=int, default=1000, help="Words per document")
parser.add_argument("--queries", type=int, default=5, help="Number of queries to run")
args = parser.parse_args()
run_benchmark(
num_documents=args.documents,
words_per_document=args.words,
num_queries=args.queries
)
if __name__ == "__main__":
main()
|