Examples

All examples are located in the GitHub repository. Their names start with the 2-digit number followed by a descriptive name. You can run examples in any order, however, if you are new to the Data Parallel Extensions for Python, it is recommended to go in the order examples are enumerated.

The following command will run the very first example of using Data Parallel Extensions for Python

python ./examples/01-hello_dpnp.py

These are listings of these examples:

EXAMPLE 01: Your first NumPy code running on GPU

import dpnp as np

x = np.asarray([1, 2, 3])
print("Array x allocated on the device:", x.device)

y = np.sum(x)

print("Result y is located on the device:", y.device)  # The same device as x
print("Shape of y is:", y.shape)  # 0-dimensional array
print("y=", y)  # Expect 6

EXAMPLE 02: Select device type while creating array

import dpnp as np

x = np.empty(3)
try:
    x = np.asarray([1, 2, 3], device="gpu")
except Exception:
    print("GPU device is not available")

print("Array x allocated on the device:", x.device)

y = np.sum(x)

print("Result y is located on the device:", y.device)  # The same device as x
print("Shape of y is:", y.shape)  # 0-dimensional array
print("y=", y)  # Expect 6

EXAMPLE 03: Compile dpnp code with numba-dpex

import dpnp as np
from numba_dpex import dpjit as njit


@njit()
def sum_it(x):  # Device queue is inferred from x. The kernel is submitted to that queue
    return np.sum(x)


x = np.empty(3)
try:
    x = np.asarray([1, 2, 3], device="gpu")
except Exception:
    print("GPU device is not available")

print("Array x allocated on the device:", x.device)

y = sum_it(x)

print("Result y is located on the device:", y.device)  # The same device as x
print("Shape of y is:", y.shape)  # 0-dimensional array
print("y=", y)  # Expect 6

EXAMPLE 04: Get information about devices

import dpctl

dpctl.lsplatform()  # Print platform information
print(
    "GPU devices:", dpctl.get_devices(device_type="gpu")
)  # Get the list of all GPU devices
print(
    "Number of GPU devices", dpctl.get_num_devices(device_type="gpu")
)  # Get the number of GPU devices
print("Has CPU devices?", dpctl.has_cpu_devices())  # Check if there are CPU devices