58 lines
2.3 KiB
Python
58 lines
2.3 KiB
Python
import os
|
|
import sys
|
|
import unittest
|
|
import torch
|
|
|
|
|
|
PROJECT_ROOT = os.path.realpath(os.path.join(os.path.dirname(__file__), "..", ".."))
|
|
MODULE_ROOT = os.path.join(PROJECT_ROOT, "modules")
|
|
sys.path.append(MODULE_ROOT)
|
|
|
|
from pytorch_wrapper.pt_unsqueeze import PtUnsqueeze # Importing the PtUnsqueeze class
|
|
|
|
|
|
class TestPtUnsqueeze(unittest.TestCase):
|
|
def setUp(self):
|
|
"""Set up test instance."""
|
|
self.node = PtUnsqueeze()
|
|
|
|
# Define test cases: (input_tensor, dim, expected_shape)
|
|
self.test_cases = [
|
|
# Case 1: Adding a singleton dimension at different positions
|
|
(torch.ones(3, 4), 0, (1, 3, 4)), # (3,4) -> (1,3,4)
|
|
(torch.ones(3, 4), 1, (3, 1, 4)), # (3,4) -> (3,1,4)
|
|
(torch.ones(3, 4), 2, (3, 4, 1)), # (3,4) -> (3,4,1)
|
|
|
|
# Case 2: Adding a singleton dimension at negative indices
|
|
(torch.ones(3, 4), -1, (3, 4, 1)), # (3,4) -> (3,4,1)
|
|
(torch.ones(3, 4), -2, (3, 1, 4)), # (3,4) -> (3,1,4)
|
|
(torch.ones(3, 4), -3, (1, 3, 4)), # (3,4) -> (1,3,4)
|
|
|
|
# Case 3: Adding multiple dimensions one by one
|
|
(torch.ones(3,), 0, (1, 3)), # (3,) -> (1,3)
|
|
(torch.ones(3,), 1, (3, 1)), # (3,) -> (3,1)
|
|
(torch.ones(3,), -1, (3, 1)), # (3,) -> (3,1)
|
|
|
|
# Case 4: Expanding an already multi-dimensional tensor
|
|
(torch.ones(2, 3, 4), 1, (2, 1, 3, 4)), # (2,3,4) -> (2,1,3,4)
|
|
(torch.ones(2, 3, 4), -1, (2, 3, 4, 1)), # (2,3,4) -> (2,3,4,1)
|
|
]
|
|
|
|
def test_unsqueeze(self):
|
|
"""Test tensor unsqueeze operation on various dimensions."""
|
|
for input_tensor, dim, expected_shape in self.test_cases:
|
|
with self.subTest(input_shape=input_tensor.shape, dim=dim):
|
|
result, = self.node.f(input_tensor, dim)
|
|
|
|
# Ensure return value is a tensor
|
|
self.assertTrue(isinstance(result, torch.Tensor), "Output is not a tensor")
|
|
|
|
# Ensure result has correct shape
|
|
self.assertEqual(result.shape, torch.Size(expected_shape), f"Expected shape {expected_shape}, got {result.shape}")
|
|
|
|
# Ensure result has the same dtype
|
|
self.assertEqual(result.dtype, input_tensor.dtype, "Dtype mismatch after unsqueeze operation")
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|