How can one interact with the LRU Cache in Python?
Python's functool module has provided functionality to interact with the LRU Cache since Python 3.2. The functool module offers a decorator that can be placed atop a Class of a function in Python. When used on functions that require large amounts of variable access and change operations, using the LRU Cache offers massive speed-up.
Example:import functools
@functools.lru_cache(maxsize = None)
def gfg():
# insert function logic here
pass
Alternatively, the maxsize can be changed to suit one's own preference. The value is measured in kbs, and maxsize takes an integer argument
Clearing LRU Cache
After the use of the cache, cache_clear() can be used for clearing or invalidating the cache.
Example 1:
import functools
@functools.lru_cache(maxsize = None)
def fib(n):
if n < 2:
return n
return fib(n-1) + fib(n-2)
fib(30)
# Before Clearing
print(fib.cache_info())
fib.cache_clear()
# After Clearing
print(fib.cache_info())
CacheInfo(hits=28, misses=31, maxsize=None, currsize=31) CacheInfo(hits=0, misses=0, maxsize=None, currsize=0)Example 2: Additionally one can also call cache_clear() from another function as well
import functools
@functools.lru_cache(maxsize = None)
def fib(n):
if n < 2:
return n
return fib(n-1) + fib(n-2)
def gfg():
fib.cache_clear()
fib(30)
# Before Clearing
print(fib.cache_info())
gfg()
# After Clearing
print(fib.cache_info())
CacheInfo(hits=28, misses=31, maxsize=None, currsize=31) CacheInfo(hits=0, misses=0, maxsize=None, currsize=0)
These methods have limitations as they are individualized, and the cache_clear() function must be typed out for each and every LRU Cache utilizing the function. We can overcome this problem, by using Python's inbuilt garbage collection module to collect all objects that have LRU Cache Wrappers, and iteratively clear each object's cache.
Example:import gc
import functools
@functools.lru_cache(maxsize = None)
def gfg():
# insert function logic here
pass
@functools.lru_cache(maxsize = None)
def gfg1():
# insert function logic here
pass
@functools.lru_cache(maxsize = None)
def gfg2():
# insert function logic here
pass
gfg()
gfg1()
gfg2()
gc.collect()
# All objects collected
objects = [i for i in gc.get_objects()
if isinstance(i, functools._lru_cache_wrapper)]
# All objects cleared
for object in objects:
object.cache_clear()