⚡️ Speed up function point_line_distance by 28%

[codeflash-ai]

📄 28% (0.28x) speedup for point_line_distance in kornia/geometry/linalg.py

⏱️ Runtime : 5.39 milliseconds → 4.22 milliseconds (best of 104 runs)

📝 Explanation and details

Changes Made.

1. In-place Operations: Used abs_() to modify numerator in-place to save time by not creating a new object.
2. Pre-calculation: Calculated denom_norm only once instead of using norm to reduce function calls.
3. Simplified Norm Calculation: Calculated the norm manually without calling norm method for efficiency. This avoids potential overheads from additional checks or support for other configurations.

✅ Correctness verification report:

Test	Status
⚙️ Existing Unit Tests	✅ 45 Passed
🌀 Generated Regression Tests	✅ 14 Passed
⏪ Replay Tests	🔘 None Found
🔎 Concolic Coverage Tests	🔘 None Found
📊 Tests Coverage	88.9%

⚙️ Existing Unit Tests Details

- geometry/test_linalg.py

🌀 Generated Regression Tests Details

from __future__ import annotations

from typing import Optional, TypeVar

# imports
import pytest  # used for our unit tests
import torch  # used for tensor operations
from kornia.core import Tensor  # used for type checking
from kornia.geometry.linalg import point_line_distance
from typing_extensions import TypeGuard

# function to test
# LICENSE HEADER MANAGED BY add-license-header
#
# Copyright 2018 Kornia Team
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#

"""The testing package contains testing-specific utilities."""

__all__ = [
    "KORNIA_CHECK_IS_TENSOR",
]

T = TypeVar("T", bound=type)
from kornia.geometry.linalg import point_line_distance


# unit tests
def test_point_line_distance_basic_2d():
    # Single 2D point and line
    point = torch.tensor([[1.0, 2.0]])
    line = torch.tensor([[1.0, -1.0, 0.0]])
    expected_distance = torch.tensor([1.0 / (2**0.5)])

def test_point_line_distance_multiple_points_single_line():
    # Multiple 2D points and a single line
    point = torch.tensor([[1.0, 2.0], [3.0, 4.0]])
    line = torch.tensor([[1.0, -1.0, 0.0]])
    expected_distances = torch.tensor([1.0 / (2**0.5), 1.0 / (2**0.5)])

def test_point_line_distance_3d_homogeneous_points():
    # Single 3D homogeneous point and line
    point = torch.tensor([[1.0, 2.0, 1.0]])
    line = torch.tensor([[1.0, -1.0, 0.0]])
    expected_distance = torch.tensor([1.0 / (2**0.5)])


def test_point_line_distance_points_on_line():
    # Points on the line
    point = torch.tensor([[1.0, 1.0]])
    line = torch.tensor([[1.0, -1.0, 0.0]])
    expected_distance = torch.tensor([0.0])

def test_point_line_distance_large_coefficients():
    # Very large coefficients
    point = torch.tensor([[1.0, 2.0]])
    line = torch.tensor([[1e10, -1e10, 1e10]])
    expected_distance = torch.tensor([1.0 / (2**0.5)])

def test_point_line_distance_invalid_input_shape():
    # Invalid shape for points
    point = torch.tensor([[1.0]])
    line = torch.tensor([[1.0, -1.0, 0.0]])
    with pytest.raises(ValueError):
        point_line_distance(point, line)

def test_point_line_distance_large_batch():
    # Large batch of points and lines
    point = torch.rand(1000, 2)
    line = torch.rand(1000, 3)
    codeflash_output = point_line_distance(point, line)
# codeflash_output is used to check that the output of the original code is the same as that of the optimized code.

import pytest  # used for our unit tests
import torch  # used to create tensors for testing
from kornia.core import Tensor  # import the Tensor type from Kornia
from kornia.geometry.linalg import point_line_distance


# unit tests
def test_single_point_line_2d():
    # Test a single point and line in 2D
    point = torch.tensor([[1.0, 1.0]])
    line = torch.tensor([[1.0, -1.0, 0.0]])  # Line: x - y = 0
    expected_distance = 0.0

def test_single_point_line_3d():
    # Test a single point and line in 3D (homogeneous coordinates)
    point = torch.tensor([[1.0, 1.0, 1.0]])
    line = torch.tensor([[1.0, -1.0, 0.0]])  # Line: x - y = 0
    expected_distance = 0.0

def test_vertical_horizontal_lines():
    # Test vertical and horizontal lines
    point = torch.tensor([[1.0, 2.0]])
    line = torch.tensor([[1.0, 0.0, -1.0]])  # Vertical line x = 1
    expected_distance = 0.0

    point = torch.tensor([[2.0, 1.0]])
    line = torch.tensor([[0.0, 1.0, -1.0]])  # Horizontal line y = 1
    expected_distance = 0.0


def test_invalid_inputs():
    # Test non-tensor inputs
    point = [[1.0, 1.0]]
    line = [1.0, -1.0, 0.0]
    with pytest.raises(TypeError):
        point_line_distance(point, line)

def test_large_batch():
    # Test large batch of points and lines
    points = torch.rand((1000, 1000, 2))
    lines = torch.rand((1000, 1000, 3))
    codeflash_output = point_line_distance(points, lines)

def test_boundary_values():
    # Test points with large coordinates
    point = torch.tensor([[1e9, 1e9]])
    line = torch.tensor([[1.0, -1.0, 0.0]])
    expected_distance = 0.0

    # Test lines with large coefficients
    point = torch.tensor([[1.0, 1.0]])
    line = torch.tensor([[1e9, -1e9, 0.0]])
    expected_distance = 0.0

def test_precision():
    # Test very close points and lines for precision
    point = torch.tensor([[1.000000001, 1.000000001]])
    line = torch.tensor([[1.0, -1.0, 0.0]])
    expected_distance = 0.0
# codeflash_output is used to check that the output of the original code is the same as that of the optimized code.

To edit these changes git checkout codeflash/optimize-point_line_distance-m8oc2zz0 and push.

[github-actions]

This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs within 7 days. Thank you for your contributions!