ttnn.SoftmaxDefaultProgramConfig

class ttnn.SoftmaxDefaultProgramConfig

Bases: pybind11_object

Default program configuration for Softmax operations.

This configuration uses the default settings for Softmax operations, providing standard behavior suitable for most use cases. It automatically selects appropriate parameters based on the input tensor characteristics.

Example

# Create input tensor
tensor = ttnn.rand((1, 1, 32, 64), dtype=ttnn.bfloat16, layout=ttnn.TILE_LAYOUT, device=device)

# Configure softmax program
compute_grid = device.compute_with_storage_grid_size()
program_config = ttnn.SoftmaxDefaultProgramConfig()

# Perform softmax
result = ttnn.softmax(tensor, dim=-1, program_config=program_config)
logger.info(f"Softmax with Program Config result: {result}")