simple profiling framwork with little overhead
Project description
KappaProfiler
Lightweight profiling utilities for identifying bottlenecks and timing program parts in your python application.
Setup
- new install:
pip install kappaprofiler
- uprade to new version:
pip install kappaprofiler --upgrade
Usage
Time your whole application
With decorators
import kappaprofiler as kp
import time
@kp.profile
def main():
time.sleep(0.3) # simulate some operation
some_method()
@kp.profile
def some_method():
time.sleep(0.5) # simulate some operation
if __name__ == "__main__":
main()
print(kp.profiler.to_string())
The result will be (time.sleep calls are not 100% accurate)
0.82 main
0.51 main.some_method
With contextmanagers
import kappaprofiler as kp
import time
def main():
with kp.named_profile("main"):
time.sleep(0.3) # simulate some operation
with kp.named_profile("method"):
some_method()
with kp.named_profile("main2"):
time.sleep(0.2) # simulate some operation
def some_method():
time.sleep(0.5) # simulate some operation
if __name__ == "__main__":
main()
print(kp.profiler.to_string())
The result will be (time.sleep calls are not 100% accurate)
0.82 main
0.51 main.method
0.20 main2
Query nodes
Each profiling entry is represented by a node from which detailed information can be retrieved
query = "main.some_method"
node = kp.profiler.get_node(query)
print(f"{query} was called {node.count} time and took {node.to_string()} seconds in total")
main.some_method was called 1 time and took 0.51 seconds in total
Time only a part of your program
import kappaprofiler as kp
with kp.Stopwatch() as sw:
# some operation
...
print(f"operation took {sw.elapsed_milliseconds} milliseconds")
print(f"operation took {sw.elapsed_seconds} seconds")
Time subparts
import kappaprofiler as kp
import time
sw1 = kp.Stopwatch()
sw2 = kp.Stopwatch()
for i in range(1, 3):
with sw1:
# operation1
time.sleep(0.1 * i)
with sw2:
# operation2
time.sleep(0.2 * i)
print(f"operation1 took {sw1.elapsed_seconds:.2f} seconds (average {sw1.average_lap_time:.2f})")
print(f"operation2 took {sw2.elapsed_seconds:.2f} seconds (average {sw2.average_lap_time:.2f})")
operation1 took 0.32 seconds (average 0.16)
operation2 took 0.61 seconds (average 0.30)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
kappaprofiler-1.0.3.tar.gz
(4.5 kB
view hashes)
Built Distribution
Close
Hashes for kappaprofiler-1.0.3-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 4212d6392f019adb63aa9763568a6a57e882755fc6b9d01ee800328bfe86b5a7 |
|
MD5 | d1e7e09bb0a1fa5a5ffa3168ee696dfc |
|
BLAKE2b-256 | 74162cfa88333674b6c144824e6f6283108dd9445e759649fc3a36feb539d1bb |