Organizing Functions with Classes

The @hog.method() decorator lets you group related functions into classes for better organization.

Basic Example

import groundhog_hpc as hog

class Statistics:
    @hog.method(endpoint="anvil")
    def compute_mean(numbers):
        import numpy as np
        return float(np.mean(numbers))

    @hog.method(endpoint="anvil")
    def compute_std(numbers):
        import numpy as np
        return float(np.std(numbers))

Note

Groundhog methods have @staticmethod semantics -- i.e. they don't receive self or cls, behaving just like standalone functions, except organized into a class.

Calling Methods

Call methods via the class or an instance:

# Via the class
mean = Statistics.compute_mean.remote([1, 2, 3, 4, 5])

# Via an instance (same behavior)
stats = Statistics()
std = stats.compute_std.remote([1, 2, 3, 4, 5])

Methods support all the same execution modes as functions: .remote(), .submit(), and .local().

When to Use Methods

Use @hog.method() when you want to organize related functions into logical groups. Use @hog.function() for standalone functions.

Complete Example

The examples/methods.py script demonstrates a Statistics class with groundhog methods.

Run it with:

hog run examples/methods.py