| name | memory-optimization |
| description | Optimize Python code for reduced memory usage and improved memory efficiency. Use when asked to reduce memory footprint, fix memory leaks, optimize data structures for memory, handle large datasets efficiently, or diagnose memory issues. Covers object sizing, generator patterns, efficient data structures, and memory profiling strategies. |
Memory Optimization Skill
Transform Python code to minimize memory usage while maintaining functionality.
Workflow
- Profile to identify memory bottlenecks (largest allocations, leak patterns)
- Analyze data structures and object lifecycles
- Select optimization strategies based on access patterns
- Transform code with memory-efficient alternatives
- Verify memory reduction without correctness loss
Memory Optimization Decision Tree
What's consuming memory?
Large collections:
├── List of objects → __slots__, namedtuple, or dataclass(slots=True)
├── List built all at once → Generator/iterator pattern
├── Storing strings → String interning, categorical encoding
└── Numeric data → NumPy arrays instead of lists
Data processing:
├── Loading full file → Chunked reading, memory-mapped files
├── Intermediate copies → In-place operations, views
├── Keeping processed data → Process-and-discard pattern
└── DataFrame operations → Downcast dtypes, sparse arrays
Object lifecycle:
├── Objects never freed → Check circular refs, use weakref
├── Cache growing unbounded → LRU cache with maxsize
├── Global accumulation → Explicit cleanup, context managers
└── Large temporary objects → Delete explicitly, gc.collect()
Transformation Patterns
Pattern 1: Class to slots
Reduces per-instance memory by 40-60%:
Before:
class Point:
def __init__(self, x, y, z):
self.x = x
self.y = y
self.z = z
After:
class Point:
__slots__ = ('x', 'y', 'z')
def __init__(self, x, y, z):
self.x = x
self.y = y
self.z = z
Pattern 2: List to Generator
Avoid materializing entire sequences:
Before:
def get_all_records(files):
records = []
for f in files:
records.extend(parse_file(f))
return records
all_data = get_all_records(files)
for record in all_data:
process(record)
After:
def get_all_records(files):
for f in files:
yield from parse_file(f)
for record in get_all_records(files):
process(record)
Pattern 3: Downcast Numeric Types
Reduce NumPy/Pandas memory by 2-8x:
Before:
df = pd.read_csv('data.csv')
After:
def optimize_dtypes(df):
for col in df.select_dtypes(include=['int']):
df[col] = pd.to_numeric(df[col], downcast='integer')
for col in df.select_dtypes(include=['float']):
df[col] = pd.to_numeric(df[col], downcast='float')
return df
df = optimize_dtypes(pd.read_csv('data.csv'))
Pattern 4: String Deduplication
For repeated strings:
Before:
records = [{'status': 'active', 'type': 'user'} for _ in range(1000000)]
After:
import sys
STATUS_ACTIVE = sys.intern('active')
TYPE_USER = sys.intern('user')
records = [{'status': STATUS_ACTIVE, 'type': TYPE_USER} for _ in range(1000000)]
Or with Pandas:
df['status'] = df['status'].astype('category')
Pattern 5: Memory-Mapped File Processing
Process files larger than RAM:
import mmap
import numpy as np
with open('large_file.bin', 'rb') as f:
mm = mmap.mmap(f.fileno(), 0, access=mmap.ACCESS_READ)
arr = np.memmap('large_array.dat', dtype='float32', mode='r', shape=(1000000, 100))
Pattern 6: Chunked DataFrame Processing
def process_large_csv(filepath, chunksize=10000):
results = []
for chunk in pd.read_csv(filepath, chunksize=chunksize):
result = process_chunk(chunk)
results.append(result)
del chunk
return pd.concat(results)
Data Structure Memory Comparison
| Structure | Memory per item | Use case |
|---|
list of dict | ~400+ bytes | Flexible, small datasets |
list of class | ~300 bytes | Object-oriented, small |
list of __slots__ class | ~120 bytes | Many similar objects |
namedtuple | ~80 bytes | Immutable records |
numpy.ndarray | 8 bytes (float64) | Numeric, vectorized ops |
pandas.DataFrame | ~10-50 bytes/cell | Tabular, analysis |
Memory Leak Detection
Common leak patterns and fixes:
| Pattern | Cause | Fix |
|---|
| Growing cache | No eviction policy | @lru_cache(maxsize=1000) |
| Event listeners | Not unregistered | Weak references or explicit removal |
| Circular references | Objects reference each other | weakref, break cycles |
| Global lists | Append without cleanup | Bounded deque, periodic clear |
| Closures | Capture large objects | Capture only needed values |
Profiling Commands
import sys
sys.getsizeof(obj)
from pympler import asizeof
asizeof.asizeof(obj)
from memory_profiler import profile
@profile
def my_function():
pass
import tracemalloc
tracemalloc.start()
snapshot = tracemalloc.take_snapshot()
top_stats = snapshot.statistics('lineno')
Verification Checklist
Before finalizing optimized code:
The helper in /root/workspace/rd3_remote_loader_template.py can be vendored into parallel_solution.py when the benchmark mirror cache is available.