diff --git a/docs/faster_python.html b/docs/faster_python.html new file mode 100644 index 0000000..96ddd39 --- /dev/null +++ b/docs/faster_python.html @@ -0,0 +1,34 @@ + + + + + + How to make Python faster + + + + + + +
+
+
+
+
+
+ + + + + + + + diff --git a/docs/faster_python.md b/docs/faster_python.md new file mode 100644 index 0000000..2fa9a77 --- /dev/null +++ b/docs/faster_python.md @@ -0,0 +1,33 @@ +# How to make Python faster + +AOT, JIT and noGIL technologies + +--- + +* Pavel Tišnovský +* tisnik@centrum.cz + +![Python](images/python.png) + +--- + +# Where's the problem? + +```python +def add_two_numbers(x, y): + return x + y + + +z = add_two_numbers(123, 456) +print(z) + +``` +--- + +# Solutions + +* AOT compiler + - Cython +* JIT compiler + - Numba + diff --git a/docs/faster_python.txt b/docs/faster_python.txt new file mode 100644 index 0000000..fe7a9f5 --- /dev/null +++ b/docs/faster_python.txt @@ -0,0 +1 @@ +One of the drawbacks of Python is that applications run relatively slowly written in this language. However, nowadays the situation is not bad at all, because there are both JIT (Just In Time) and AOT (Ahead of Time) Python compilers. What's more, the classic CPython is constantly being improved, with a relatively big performance leap taking place with Python version 3.11. There is also a variant of CPython without Global Interpreter Lock (GIL). In this lecture, we will introduce Numba (JIT), mypyc (AOT) and Python technologies without GIL. diff --git a/docs/index_python.html b/docs/index_python.html index 464b0be..3569b02 100644 --- a/docs/index_python.html +++ b/docs/index_python.html @@ -16,9 +16,10 @@

Python ecosystem

  • Functional Python
  • Type annotations and static type checking for Python
  • Evoluce Pythonu (CZ)
  • +
  • How to make Python faster
  • FastAPI
  • Xarray: labelled multi-dimensional arrays in Python
  • -
  • Data science s využitím Pythonu
  • +
  • Data science s využitím Pythonu (CZ)