chromium/third_party/blink/web_tests/platform/mac/virtual/webnn-service-with-gpu/external/wpt/webnn/conformance_tests/identity.https.any.worker_gpu-expected.txt

This is a testharness.js-based test.
[FAIL] identity float32 0D scalar
  assert_true: assert_array_approx_equals_ulp: test identity float32 actual -4.2734375 should be close enough to expected -4.273642539978027 by the acceptable 0 ULP distance, but they have 430 ULP distance expected true got false
[FAIL] identity float32 1D constant tensor
  assert_true: assert_array_approx_equals_ulp: test identity float32 actual 0.37744140625 should be close enough to expected 0.377551406621933 by the acceptable 0 ULP distance, but they have 3691 ULP distance expected true got false
[FAIL] identity float32 1D tensor
  assert_true: assert_array_approx_equals_ulp: test identity float32 actual 0.37744140625 should be close enough to expected 0.377551406621933 by the acceptable 0 ULP distance, but they have 3691 ULP distance expected true got false
[FAIL] identity float32 2D tensor
  assert_true: assert_array_approx_equals_ulp: test identity float32 actual 0.37744140625 should be close enough to expected 0.377551406621933 by the acceptable 0 ULP distance, but they have 3691 ULP distance expected true got false
[FAIL] identity float32 3D tensor
  assert_true: assert_array_approx_equals_ulp: test identity float32 actual 0.37744140625 should be close enough to expected 0.377551406621933 by the acceptable 0 ULP distance, but they have 3691 ULP distance expected true got false
[FAIL] identity float32 4D tensor
  assert_true: assert_array_approx_equals_ulp: test identity float32 actual 0.37744140625 should be close enough to expected 0.377551406621933 by the acceptable 0 ULP distance, but they have 3691 ULP distance expected true got false
[FAIL] identity float32 5D tensor
  assert_true: assert_array_approx_equals_ulp: test identity float32 actual 0.37744140625 should be close enough to expected 0.377551406621933 by the acceptable 0 ULP distance, but they have 3691 ULP distance expected true got false
Harness: the test ran to completion.