FF...................................................................... [ 97%]
..                                                                       [100%]
=================================== FAILURES ===================================
___ test_ordinary_weights_forward_and_actual_update_match_reusable_baseline ____

tmp_path = PosixPath('/tmp/pytest-of-root/pytest-0/test_ordinary_weights_forward_0')

    def test_ordinary_weights_forward_and_actual_update_match_reusable_baseline(tmp_path):
        spec = options("D4")
        old_spec = {**spec, "arm": "standard", "base_unique_blocks": 4}
>       old, model = old_construct(old_spec, "cpu"), ms.construct(spec["model"], spec, "cpu")

tests/test_moe_sequential.py:56: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

spec = {'architecture': 'D4', 'arm': 'standard', 'balance_coefficient': 0.01, 'base_unique_blocks': 4, ...}
device = 'cpu'

    def construct(spec, device):
        arm = spec["arm"]
        if arm not in ARMS:
            raise ValueError(f"Unknown architecture: {arm}")
        base, repeats = int(spec["unique_layers"]), int(spec["repeats"])
        if base != int(spec["base_unique_blocks"]):
            raise ValueError("Stored and base layer counts must agree")
        if arm == "standard" and repeats != 1:
            raise ValueError("The ordinary baseline executes independent layers once")
        if arm != "standard" and repeats < 2:
            raise ValueError("Loop comparison requires at least two rounds")
>       reference = int(spec["initialization_reference_layers"])
E       KeyError: 'initialization_reference_layers'

src/llm_memory_editability/loop_kv_learning.py:86: KeyError
_____ test_sparse_total_and_active_parameters_have_separate_dense_controls _____

    def test_sparse_total_and_active_parameters_have_separate_dense_controls():
        ledgers = {
            arm: ms.manifest(ms.construct(options(arm)["model"], options(arm), "cpu"))
            for arm in ms.ARMS
        }
        assert ledgers["M4"]["unique_parameters"] > ledgers["D4"]["unique_parameters"]
>       assert (
            abs(ledgers["M4"]["unique_parameters"] - ledgers["W4"]["unique_parameters"])
            < 0.02 * ledgers["M4"]["unique_parameters"]
        )
E       assert 960 < (0.02 * 40656)
E        +  where 960 = abs((40656 - 39696))

tests/test_moe_sequential.py:74: AssertionError
=============================== warnings summary ===============================
../../../opt/conda/lib/python3.10/site-packages/urllib3/util/ssl_.py:281
  /opt/conda/lib/python3.10/site-packages/urllib3/util/ssl_.py:281: DeprecationWarning: ssl.PROTOCOL_TLS is deprecated
    context = SSLContext(ssl_version or PROTOCOL_TLS)

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
=========================== short test summary info ============================
FAILED tests/test_moe_sequential.py::test_ordinary_weights_forward_and_actual_update_match_reusable_baseline
FAILED tests/test_moe_sequential.py::test_sparse_total_and_active_parameters_have_separate_dense_controls
2 failed, 72 passed, 1 warning in 8.69s