Parakeet is a runtime accelerator for an array-oriented subset of Python. If you're doing a lot of number crunching in Python, Parakeet may be able to significantly speed up your code.
To accelerate a function, wrap it with Parakeet's @jit decorator:
import numpy as np
from parakeet import jit
x = np.array([1,2,3])
y = np.tanh(x) * alpha + beta
@jit
def fast(x, alpha = 0.5, beta = 0.3):
return np.tanh(x) * alpha + beta
@jit
def loopy(x, alpha = 0.5, beta = 0.3):
y = np.empty_like(x)
for i in xrange(len(x)):
y[i] = np.tanh(x[i] * alpha + beta)
return y
@jit
def comprehension(x, alpha = 0.5, beta = 0.3):
return np.array([np.tanh(xi*alpha + beta) for xi in x])
assert fast(x) == y
assert loopy(x) == y
assert comprehension(x) == y
You should be able to install Parakeet from its PyPI package by running:
pip install parakeet
Parakeet is written for Python 2.7 (sorry internet) and depends on:
- treelike
- nose for unit tests
- NumPy and SciPy
Optional (if using the LLVM backend):
Parakeet cannot accelerate arbitrary Python code, it only supports a limited subset of the language:
- Scalar operations (i.e. addition, multiplication, etc...)
- Control flow (if-statements, loops, etc...)
- Tuples
- Slices
- NumPy arrays (and some NumPy library functions)
- List literals (interpreted as array construction)
- List comprehensions (interpreted as array comprehensions)
- Parakeet's "adverbs" (higher order array operations like parakeet.map, parakeet.reduce)
Your untyped function gets used as a template from which multiple type specializations are generated (for each distinct set of input types). These typed functions are then churned through many optimizations before finally getting translated into native code. For more information about the project you can watch the Parakeet presentation from this year's PyData Boston, look at the HotPar slides from last year or contact the Alex Rubinsteyn.
Parakeet currently supports compilation to C or LLVM. To switch between these options change parakeet.config.default_backend
to either "c" or "llvm".