hsdpy 0.1.0

HsdPy

HsdPy library allows users to use Hsdlib in Python.

Installation

pip install hsdpy

Examples

Below is a simple usage example of HsdPy:

import hsdpy

# HsdPy uses NumPy for array handling
import numpy as np

# Create two NumPy arrays of float32 type
a = np.array([1.0, 2.0, 3.0], dtype=np.float32)
b = np.array([4.0, 5.0, 6.0], dtype=np.float32)

# Calculate the euclidean distance between the two arrays
dist_euc = hsdpy.dist_sqeuclidean_f32(a, b) ** 0.5

# Calculate Manhattan distance
dist_man = hsdpy.dist_manhattan_f32(a, b)

# Let's see the results
print(f"Euclidean distance: {dist_euc}")  # 5.196
print(f"Manhattan distance: {dist_man}")  # 9.0

# See the SIMD backend in use
print(f"Backend: {hsdpy.get_backend()}")

Check out hsdpy_example.py for more detailed examples.

API Summary

Notes

• HsdPy provides the HsdError exception class for error handling. It is a custom exception class wraps the Hsdlib error codes to make them more Pythonic.
• All distance and similarity functions expect one-dimensional NumPy arrays as input.
• Functions will raise NotImplementedError if the corresponding Hsdlib function is not implemented for the given data type.

License

HsdPy is licensed under the MIT License.