atomscale 0.18.0__tar.gz → 0.19.0__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {atomscale-0.18.0 → atomscale-0.19.0}/PKG-INFO +1 -1
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/guides/analysis-results.rst +34 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/client.py +86 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/group.py +10 -1
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/__init__.py +6 -0
- atomscale-0.19.0/src/atomscale/timeseries/physical_sample.py +137 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale.egg-info/PKG-INFO +1 -1
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale.egg-info/SOURCES.txt +2 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale.egg-info/scm_file_list.json +2 -0
- atomscale-0.19.0/src/atomscale.egg-info/scm_version.json +8 -0
- atomscale-0.19.0/tests/test_physical_sample_timeseries.py +290 -0
- atomscale-0.18.0/src/atomscale.egg-info/scm_version.json +0 -8
- {atomscale-0.18.0 → atomscale-0.19.0}/.github/workflows/release.yml +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/.github/workflows/testing.yml +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/.github/workflows/upgrade_dependencies.yml +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/.gitignore +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/.pre-commit-config.yaml +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/CHANGELOG.md +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/LICENSE +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/MANIFEST.in +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/README.md +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/atomicds-shim-dist/pyproject.toml +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/Makefile +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/_static/AtomscaleLogoFull.png +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/_static/custom.css +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/_templates/custom-class-template.rst +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/_templates/custom-module-template.rst +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/conf.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/guides/monitor-live.rst +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/guides/quickstart.rst +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/guides/search-and-download.rst +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/guides/stream-rheed.rst +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/guides/stream-timeseries.rst +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/guides/upload-files.rst +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/index.rst +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/make.bat +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/docs/modules.rst +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/examples/changepoints.ipynb +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/examples/general_use.ipynb +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/examples/rheed_streaming.ipynb +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/examples/timeseries_polling.ipynb +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/examples/vxwse2-placeholder/task1_films.ipynb +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/examples/vxwse2-placeholder/task1_sapphire.ipynb +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/examples/vxwse2-placeholder/task2_composition.ipynb +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/pyproject.toml +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-macos-latest_py3.10.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-macos-latest_py3.10_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-macos-latest_py3.11.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-macos-latest_py3.11_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-macos-latest_py3.12.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-macos-latest_py3.12_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-macos-latest_py3.13.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-macos-latest_py3.13_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-ubuntu-latest_py3.10.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-ubuntu-latest_py3.10_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-ubuntu-latest_py3.11.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-ubuntu-latest_py3.11_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-ubuntu-latest_py3.12.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-ubuntu-latest_py3.12_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-ubuntu-latest_py3.13.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-ubuntu-latest_py3.13_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-windows-latest_py3.10.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-windows-latest_py3.10_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-windows-latest_py3.11.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-windows-latest_py3.11_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-windows-latest_py3.12.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-windows-latest_py3.12_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-windows-latest_py3.13.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/requirements/requirements-windows-latest_py3.13_extras.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/scripts/ci_auth_preflight.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/setup.cfg +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomicds/__init__.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/__init__.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/core/__init__.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/core/client.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/core/files.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/core/utils.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/__init__.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/changepoint.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/ellipsometry.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/embeddings.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/metrology.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/optical.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/photoluminescence.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/raman.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/recipe.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/rheed_image.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/rheed_video.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/similarity_trajectory.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/tool_state.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/unknown.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/xps.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/results/xrd.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/similarity/__init__.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/similarity/embedding_provider.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/similarity/polling.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/similarity/provider.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/streaming/Cargo.lock +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/streaming/Cargo.toml +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/streaming/__init__.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/streaming/rheed_stream.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/streaming/rheed_stream.pyi +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/streaming/src/initialize.rs +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/streaming/src/lib.rs +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/streaming/src/timeseries.rs +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/streaming/src/upload.rs +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/streaming/src/utils.rs +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/_polling_utils.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/align.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/ellipsometry.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/metrology.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/optical.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/polling.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/provider.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/recipe.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/registry.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/rheed.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/sample.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale/timeseries/tool_state.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale.egg-info/dependency_links.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale.egg-info/requires.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/src/atomscale.egg-info/top_level.txt +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/__init__.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/_mock_http_server.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/conftest.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/data/test_rheed.mp4 +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_align.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_atomicds_alias.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_changepoint.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_client.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_core.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_ellipsometry.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_embeddings.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_metrology.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_optical.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_photoluminescence.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_polling.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_raman.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_recipe.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_rheed_embedding.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_rheed_image.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_rheed_stream.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_rheed_timeseries.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_rheed_video.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_similarity_matches.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_similarity_trajectory.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_timeseries_stream.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/tests/test_xps.py +0 -0
- {atomscale-0.18.0 → atomscale-0.19.0}/uv.lock +0 -0
|
@@ -97,6 +97,40 @@ number — with :meth:`~atomscale.client.Client.get_frame_masks`:
|
|
|
97
97
|
|
|
98
98
|
masks = client.get_frame_masks(data_id, decode=True) # {frame_number: (H, W) array}
|
|
99
99
|
|
|
100
|
+
Sample-Level Results
|
|
101
|
+
--------------------
|
|
102
|
+
|
|
103
|
+
Some results are computed per *physical sample* rather than per data item —
|
|
104
|
+
``rheed_quality`` and ``composition_metric`` are compiled from a sample's
|
|
105
|
+
constituent RHEED videos into one series each. Fetch them with
|
|
106
|
+
:meth:`~atomscale.client.Client.get_physical_sample_timeseries`:
|
|
107
|
+
|
|
108
|
+
.. code-block:: python
|
|
109
|
+
|
|
110
|
+
ts = client.get_physical_sample_timeseries(
|
|
111
|
+
physical_sample_id, property_names=["rheed_quality"]
|
|
112
|
+
)
|
|
113
|
+
|
|
114
|
+
# Long form: one row per (property, sample-point). Reduce to a scalar yourself.
|
|
115
|
+
q = ts.loc[ts.property_name == "rheed_quality", "value"].dropna()
|
|
116
|
+
print(q.mean())
|
|
117
|
+
|
|
118
|
+
The frame is long (not wide) because distinct properties can have different
|
|
119
|
+
axes, so a wide join on ``real_time_seconds`` would mis-align them. ``value`` is
|
|
120
|
+
``NaN`` for gaps; provenance columns (``result_id``, ``last_updated``,
|
|
121
|
+
``generating_dbos_workflow_id``) travel alongside, and per-property
|
|
122
|
+
``constituent_data_ids`` are in ``ts.attrs["constituent_data_ids"]``.
|
|
123
|
+
``property_names`` filters client-side; omit it for all properties.
|
|
124
|
+
|
|
125
|
+
The same metrics are attached to :meth:`~atomscale.client.Client.get_physical_sample`
|
|
126
|
+
results as ``sample_metrics`` (pass ``include_sample_metrics=False`` to skip the
|
|
127
|
+
extra request):
|
|
128
|
+
|
|
129
|
+
.. code-block:: python
|
|
130
|
+
|
|
131
|
+
sample = client.get_physical_sample(physical_sample_id)
|
|
132
|
+
print(sample.sample_metrics) # None if the sample has no computed metrics
|
|
133
|
+
|
|
100
134
|
Embedding Vectors
|
|
101
135
|
-----------------
|
|
102
136
|
|
|
@@ -41,6 +41,9 @@ from atomscale.results import (
|
|
|
41
41
|
)
|
|
42
42
|
from atomscale.results.group import PhysicalSampleResult, ProjectResult
|
|
43
43
|
from atomscale.timeseries.align import align_timeseries
|
|
44
|
+
from atomscale.timeseries.physical_sample import (
|
|
45
|
+
physical_sample_timeseries_to_dataframe,
|
|
46
|
+
)
|
|
44
47
|
from atomscale.timeseries.registry import get_provider
|
|
45
48
|
|
|
46
49
|
TimeseriesDomain = Literal["rheed", "optical", "tool_state"]
|
|
@@ -1066,6 +1069,7 @@ class Client(BaseClient):
|
|
|
1066
1069
|
*,
|
|
1067
1070
|
include_organization_data: bool = True,
|
|
1068
1071
|
align: bool | str = False,
|
|
1072
|
+
include_sample_metrics: bool = True,
|
|
1069
1073
|
) -> PhysicalSampleResult:
|
|
1070
1074
|
"""Get all data for a physical sample.
|
|
1071
1075
|
|
|
@@ -1073,6 +1077,11 @@ class Client(BaseClient):
|
|
|
1073
1077
|
physical_sample_id: Identifier of the physical sample.
|
|
1074
1078
|
include_organization_data: Whether to include organization data. Defaults to True.
|
|
1075
1079
|
align: Whether to align timeseries data. If truthy, an aligned DataFrame is returned.
|
|
1080
|
+
include_sample_metrics: Whether to also fetch the sample-scoped computed
|
|
1081
|
+
timeseries results (``rheed_quality``, ``composition_metric``, …) and
|
|
1082
|
+
expose them on ``PhysicalSampleResult.sample_metrics``. Defaults to True.
|
|
1083
|
+
This is one extra request; set ``False`` to skip it. See
|
|
1084
|
+
:meth:`get_physical_sample_timeseries`.
|
|
1076
1085
|
"""
|
|
1077
1086
|
physical_samples: list[dict] | None = self._get( # type: ignore # noqa: PGH003
|
|
1078
1087
|
sub_url="physical_samples/",
|
|
@@ -1113,12 +1122,89 @@ class Client(BaseClient):
|
|
|
1113
1122
|
else None
|
|
1114
1123
|
)
|
|
1115
1124
|
sample_id = physical_sample_id
|
|
1125
|
+
|
|
1126
|
+
# Sample-scoped computed metrics live in a separate table reached by a
|
|
1127
|
+
# dedicated endpoint (not part of the per-data-item curated join above).
|
|
1128
|
+
# This is best-effort enrichment: it is default-enabled and must never
|
|
1129
|
+
# abort an otherwise-successful sample fetch (nor a ``get_project``
|
|
1130
|
+
# aggregation that loops over samples). A 404 (``_get`` -> None) simply
|
|
1131
|
+
# yields no metrics; any other HTTP error or transport failure is caught
|
|
1132
|
+
# and surfaced as a warning, leaving ``sample_metrics`` None. Callers who
|
|
1133
|
+
# need the metrics to fail loudly use :meth:`get_physical_sample_timeseries`.
|
|
1134
|
+
sample_metrics: DataFrame | None = None
|
|
1135
|
+
if include_sample_metrics:
|
|
1136
|
+
try:
|
|
1137
|
+
raw_metrics = self._get(
|
|
1138
|
+
sub_url=f"physical_samples/{physical_sample_id}/timeseries/"
|
|
1139
|
+
)
|
|
1140
|
+
if raw_metrics is not None:
|
|
1141
|
+
sample_metrics = physical_sample_timeseries_to_dataframe(
|
|
1142
|
+
raw_metrics
|
|
1143
|
+
) # type: ignore[arg-type]
|
|
1144
|
+
except (ClientError, RequestException) as exc:
|
|
1145
|
+
warnings.warn(
|
|
1146
|
+
f"Could not fetch sample metrics for physical sample "
|
|
1147
|
+
f"'{physical_sample_id}': {exc}. Returning the sample without "
|
|
1148
|
+
"them; call get_physical_sample_timeseries(...) directly to see "
|
|
1149
|
+
"the error, or pass include_sample_metrics=False to skip this.",
|
|
1150
|
+
stacklevel=2,
|
|
1151
|
+
)
|
|
1152
|
+
|
|
1116
1153
|
return PhysicalSampleResult(
|
|
1117
1154
|
physical_sample_id=sample_id,
|
|
1118
1155
|
physical_sample_name=sample_name,
|
|
1119
1156
|
data_results=results,
|
|
1120
1157
|
aligned_timeseries=ts_aligned,
|
|
1121
1158
|
non_timeseries=non_timeseries,
|
|
1159
|
+
sample_metrics=sample_metrics,
|
|
1160
|
+
)
|
|
1161
|
+
|
|
1162
|
+
def get_physical_sample_timeseries(
|
|
1163
|
+
self,
|
|
1164
|
+
physical_sample_id: str,
|
|
1165
|
+
*,
|
|
1166
|
+
property_names: list[str] | None = None,
|
|
1167
|
+
) -> DataFrame:
|
|
1168
|
+
"""Current sample-scoped computed timeseries results for a physical sample.
|
|
1169
|
+
|
|
1170
|
+
Returns the latest row per property (``rheed_quality``,
|
|
1171
|
+
``composition_metric``, …) from the ``physical_sample_timeseries_results``
|
|
1172
|
+
table — the outputs of per-sample DBOS workflows, content-addressed by
|
|
1173
|
+
their constituent ``data_id`` set. These are the headline "sample result,"
|
|
1174
|
+
distinct from :meth:`get_physical_sample`'s ``aligned_timeseries``, which
|
|
1175
|
+
only re-joins each constituent data item's own curated RHEED series.
|
|
1176
|
+
|
|
1177
|
+
The series are returned faithfully (no scalar reduction): callers wanting a
|
|
1178
|
+
single number reduce the returned series themselves.
|
|
1179
|
+
|
|
1180
|
+
Args:
|
|
1181
|
+
physical_sample_id: Identifier of the physical sample.
|
|
1182
|
+
property_names: Restrict the result to these properties, e.g.
|
|
1183
|
+
``["rheed_quality"]``. Filtered client-side (the endpoint has no
|
|
1184
|
+
filter param yet). ``None`` (default) returns all properties.
|
|
1185
|
+
|
|
1186
|
+
Returns:
|
|
1187
|
+
DataFrame: Long form — one row per (property, sample-point) — with
|
|
1188
|
+
columns ``property_name``, ``real_time_seconds``, ``value`` (float,
|
|
1189
|
+
``NaN`` for gaps), ``result_id``, ``last_updated``, and
|
|
1190
|
+
``generating_dbos_workflow_id``. Long (not wide) because distinct
|
|
1191
|
+
properties can have different axes, so a wide join on
|
|
1192
|
+
``real_time_seconds`` would mis-align them. Per-property
|
|
1193
|
+
``constituent_data_ids`` are in ``df.attrs["constituent_data_ids"]``.
|
|
1194
|
+
A sample with no sample-scoped metrics yields an empty DataFrame.
|
|
1195
|
+
|
|
1196
|
+
Raises:
|
|
1197
|
+
ClientError: If the physical sample does not exist (404).
|
|
1198
|
+
"""
|
|
1199
|
+
raw = self._get(sub_url=f"physical_samples/{physical_sample_id}/timeseries/")
|
|
1200
|
+
if raw is None:
|
|
1201
|
+
raise ClientError(
|
|
1202
|
+
f"No physical sample found with id '{physical_sample_id}' "
|
|
1203
|
+
"(physical_samples/{id}/timeseries/ returned 404).",
|
|
1204
|
+
status_code=404,
|
|
1205
|
+
)
|
|
1206
|
+
return physical_sample_timeseries_to_dataframe( # type: ignore[arg-type]
|
|
1207
|
+
raw, property_names=property_names
|
|
1122
1208
|
)
|
|
1123
1209
|
|
|
1124
1210
|
def get_project(
|
|
@@ -14,13 +14,22 @@ class PhysicalSampleResult(MSONable):
|
|
|
14
14
|
data_results: list,
|
|
15
15
|
aligned_timeseries: DataFrame | None,
|
|
16
16
|
non_timeseries: list,
|
|
17
|
+
sample_metrics: DataFrame | None = None,
|
|
17
18
|
):
|
|
18
|
-
"""Aggregated results for a physical sample.
|
|
19
|
+
"""Aggregated results for a physical sample.
|
|
20
|
+
|
|
21
|
+
``sample_metrics`` holds the sample-scoped computed timeseries results
|
|
22
|
+
(``rheed_quality``, ``composition_metric``, …) in the long form returned
|
|
23
|
+
by :meth:`~atomscale.client.Client.get_physical_sample_timeseries`, or
|
|
24
|
+
``None`` when they were not fetched. It is distinct from
|
|
25
|
+
``aligned_timeseries``, which is the per-data-item curated join.
|
|
26
|
+
"""
|
|
19
27
|
self.physical_sample_id = physical_sample_id
|
|
20
28
|
self.physical_sample_name = physical_sample_name
|
|
21
29
|
self.data_results = data_results
|
|
22
30
|
self.aligned_timeseries = aligned_timeseries
|
|
23
31
|
self.non_timeseries = non_timeseries
|
|
32
|
+
self.sample_metrics = sample_metrics
|
|
24
33
|
|
|
25
34
|
|
|
26
35
|
class ProjectResult(MSONable):
|
|
@@ -22,6 +22,10 @@ from .align import align_timeseries
|
|
|
22
22
|
from .ellipsometry import EllipsometryProvider
|
|
23
23
|
from .metrology import MetrologyProvider
|
|
24
24
|
from .optical import OpticalProvider
|
|
25
|
+
from .physical_sample import (
|
|
26
|
+
PHYSICAL_SAMPLE_TS_COLUMNS,
|
|
27
|
+
physical_sample_timeseries_to_dataframe,
|
|
28
|
+
)
|
|
25
29
|
from .polling import (
|
|
26
30
|
aiter_poll,
|
|
27
31
|
iter_poll,
|
|
@@ -35,6 +39,7 @@ from .rheed import RHEEDProvider
|
|
|
35
39
|
from .tool_state import ToolStateProvider
|
|
36
40
|
|
|
37
41
|
__all__ = [
|
|
42
|
+
"PHYSICAL_SAMPLE_TS_COLUMNS",
|
|
38
43
|
"EllipsometryProvider",
|
|
39
44
|
"MetrologyProvider",
|
|
40
45
|
"OpticalProvider",
|
|
@@ -46,6 +51,7 @@ __all__ = [
|
|
|
46
51
|
"align_timeseries",
|
|
47
52
|
"get_provider",
|
|
48
53
|
"iter_poll",
|
|
54
|
+
"physical_sample_timeseries_to_dataframe",
|
|
49
55
|
"properties_payload_to_dataframe",
|
|
50
56
|
"start_polling_task",
|
|
51
57
|
"start_polling_thread",
|
|
@@ -0,0 +1,137 @@
|
|
|
1
|
+
"""Parse sample-scoped computed timeseries results.
|
|
2
|
+
|
|
3
|
+
The backend endpoint ``GET /physical_samples/{id}/timeseries/`` returns the
|
|
4
|
+
*current* row per computed property that per-sample DBOS workflows produce —
|
|
5
|
+
today ``rheed_quality`` and ``composition_metric``. Each property is a compiled
|
|
6
|
+
series content-addressed by its constituent ``data_id`` set; the endpoint
|
|
7
|
+
returns the latest row for each property.
|
|
8
|
+
|
|
9
|
+
This module converts that payload into a long-form DataFrame. Long form (rather
|
|
10
|
+
than a wide join on ``real_time_seconds``) because distinct properties can carry
|
|
11
|
+
*different* axes — different ``result_id`` / constituent-data-id sets — so a
|
|
12
|
+
naive wide join would silently mis-align them.
|
|
13
|
+
"""
|
|
14
|
+
|
|
15
|
+
from __future__ import annotations
|
|
16
|
+
|
|
17
|
+
from collections.abc import Mapping, Sequence
|
|
18
|
+
from typing import Any
|
|
19
|
+
|
|
20
|
+
import numpy as np
|
|
21
|
+
import pandas as pd
|
|
22
|
+
from pandas import DataFrame
|
|
23
|
+
|
|
24
|
+
# Long-form output schema (scalar columns). Per-property list-valued provenance
|
|
25
|
+
# (``constituent_data_ids``) is carried in ``DataFrame.attrs`` instead of a
|
|
26
|
+
# list-in-cell column so the frame stays flat and groupby/filter-friendly.
|
|
27
|
+
PHYSICAL_SAMPLE_TS_COLUMNS: tuple[str, ...] = (
|
|
28
|
+
"property_name",
|
|
29
|
+
"real_time_seconds",
|
|
30
|
+
"value",
|
|
31
|
+
"result_id",
|
|
32
|
+
"last_updated",
|
|
33
|
+
"generating_dbos_workflow_id",
|
|
34
|
+
)
|
|
35
|
+
|
|
36
|
+
|
|
37
|
+
def _float_array(values: Sequence[Any]) -> np.ndarray:
|
|
38
|
+
"""Coerce a sequence to ``float64``, mapping ``None`` (JSON null) to ``NaN``."""
|
|
39
|
+
return np.array(
|
|
40
|
+
[np.nan if v is None else float(v) for v in values], dtype="float64"
|
|
41
|
+
)
|
|
42
|
+
|
|
43
|
+
|
|
44
|
+
def _empty_frame() -> DataFrame:
|
|
45
|
+
"""Empty long-form frame with the standard columns and dtypes."""
|
|
46
|
+
frame = DataFrame({c: [] for c in PHYSICAL_SAMPLE_TS_COLUMNS})
|
|
47
|
+
frame["real_time_seconds"] = frame["real_time_seconds"].astype("float64")
|
|
48
|
+
frame["value"] = frame["value"].astype("float64")
|
|
49
|
+
frame["last_updated"] = pd.to_datetime(frame["last_updated"])
|
|
50
|
+
frame.attrs["constituent_data_ids"] = {}
|
|
51
|
+
return frame
|
|
52
|
+
|
|
53
|
+
|
|
54
|
+
def physical_sample_timeseries_to_dataframe(
|
|
55
|
+
payload: Mapping[str, Any],
|
|
56
|
+
*,
|
|
57
|
+
property_names: Sequence[str] | None = None,
|
|
58
|
+
) -> DataFrame:
|
|
59
|
+
"""Convert a ``PhysicalSampleTimeseriesResponse`` payload to long form.
|
|
60
|
+
|
|
61
|
+
Args:
|
|
62
|
+
payload: The decoded response body, with a ``"properties"`` list whose
|
|
63
|
+
entries carry ``property_name``, ``property_values`` (JSON ``null``
|
|
64
|
+
for gaps), ``real_time_seconds`` (1:1 with values), and provenance
|
|
65
|
+
(``result_id``, ``last_updated``, ``constituent_data_ids``,
|
|
66
|
+
``generating_dbos_workflow_id``). If ``unix_times`` is present (a
|
|
67
|
+
backend follow-up), it is passed through as a column.
|
|
68
|
+
property_names: If given, keep only these properties (client-side; the
|
|
69
|
+
endpoint has no filter param yet). ``None`` keeps all.
|
|
70
|
+
|
|
71
|
+
Returns:
|
|
72
|
+
DataFrame: One row per (property, sample-point) with the columns in
|
|
73
|
+
:data:`PHYSICAL_SAMPLE_TS_COLUMNS` (plus ``unix_times`` when the payload
|
|
74
|
+
provides it). ``value`` and ``real_time_seconds`` are ``float64`` with
|
|
75
|
+
``NaN`` for null gaps; ``last_updated`` is datetime. Per-property
|
|
76
|
+
``constituent_data_ids`` are stored in ``df.attrs["constituent_data_ids"]``
|
|
77
|
+
keyed by property name. Empty / fully-filtered payloads yield an empty
|
|
78
|
+
frame (never an error).
|
|
79
|
+
|
|
80
|
+
Raises:
|
|
81
|
+
ValueError: If a property's ``property_values`` and ``real_time_seconds``
|
|
82
|
+
differ in length (malformed backend response).
|
|
83
|
+
"""
|
|
84
|
+
properties = list(payload.get("properties") or [])
|
|
85
|
+
|
|
86
|
+
if property_names is not None:
|
|
87
|
+
wanted = set(property_names)
|
|
88
|
+
properties = [p for p in properties if p.get("property_name") in wanted]
|
|
89
|
+
|
|
90
|
+
if not properties:
|
|
91
|
+
return _empty_frame()
|
|
92
|
+
|
|
93
|
+
has_unix = any(p.get("unix_times") for p in properties)
|
|
94
|
+
|
|
95
|
+
frames: list[DataFrame] = []
|
|
96
|
+
constituent: dict[str, list] = {}
|
|
97
|
+
for prop in properties:
|
|
98
|
+
name = prop.get("property_name")
|
|
99
|
+
values = list(prop.get("property_values") or [])
|
|
100
|
+
axis = list(prop.get("real_time_seconds") or [])
|
|
101
|
+
if len(values) != len(axis):
|
|
102
|
+
raise ValueError(
|
|
103
|
+
f"Malformed physical-sample timeseries for property {name!r}: "
|
|
104
|
+
f"{len(values)} value(s) vs {len(axis)} time point(s)."
|
|
105
|
+
)
|
|
106
|
+
|
|
107
|
+
axis_arr = _float_array(axis)
|
|
108
|
+
data: dict[str, Any] = {
|
|
109
|
+
# Scalars broadcast to the axis length (0 rows when the axis is empty).
|
|
110
|
+
"property_name": name,
|
|
111
|
+
"real_time_seconds": axis_arr,
|
|
112
|
+
"value": _float_array(values),
|
|
113
|
+
"result_id": prop.get("result_id"),
|
|
114
|
+
"last_updated": prop.get("last_updated"),
|
|
115
|
+
"generating_dbos_workflow_id": prop.get("generating_dbos_workflow_id"),
|
|
116
|
+
}
|
|
117
|
+
if has_unix:
|
|
118
|
+
unix = list(prop.get("unix_times") or [])
|
|
119
|
+
data["unix_times"] = (
|
|
120
|
+
_float_array(unix)
|
|
121
|
+
if len(unix) == len(axis_arr)
|
|
122
|
+
else np.full(len(axis_arr), np.nan, dtype="float64")
|
|
123
|
+
)
|
|
124
|
+
frames.append(DataFrame(data))
|
|
125
|
+
constituent[name] = list(prop.get("constituent_data_ids") or [])
|
|
126
|
+
|
|
127
|
+
out = pd.concat(frames, ignore_index=True)
|
|
128
|
+
|
|
129
|
+
ordered = ["property_name", "real_time_seconds"]
|
|
130
|
+
if has_unix:
|
|
131
|
+
ordered.append("unix_times")
|
|
132
|
+
ordered += ["value", "result_id", "last_updated", "generating_dbos_workflow_id"]
|
|
133
|
+
out = out[ordered]
|
|
134
|
+
|
|
135
|
+
out["last_updated"] = pd.to_datetime(out["last_updated"], errors="coerce")
|
|
136
|
+
out.attrs["constituent_data_ids"] = constituent
|
|
137
|
+
return out
|
|
@@ -109,6 +109,7 @@ src/atomscale/timeseries/align.py
|
|
|
109
109
|
src/atomscale/timeseries/ellipsometry.py
|
|
110
110
|
src/atomscale/timeseries/metrology.py
|
|
111
111
|
src/atomscale/timeseries/optical.py
|
|
112
|
+
src/atomscale/timeseries/physical_sample.py
|
|
112
113
|
src/atomscale/timeseries/polling.py
|
|
113
114
|
src/atomscale/timeseries/provider.py
|
|
114
115
|
src/atomscale/timeseries/recipe.py
|
|
@@ -129,6 +130,7 @@ tests/test_embeddings.py
|
|
|
129
130
|
tests/test_metrology.py
|
|
130
131
|
tests/test_optical.py
|
|
131
132
|
tests/test_photoluminescence.py
|
|
133
|
+
tests/test_physical_sample_timeseries.py
|
|
132
134
|
tests/test_polling.py
|
|
133
135
|
tests/test_raman.py
|
|
134
136
|
tests/test_recipe.py
|
|
@@ -103,6 +103,7 @@
|
|
|
103
103
|
"src/atomscale/timeseries/ellipsometry.py",
|
|
104
104
|
"src/atomscale/timeseries/metrology.py",
|
|
105
105
|
"src/atomscale/timeseries/optical.py",
|
|
106
|
+
"src/atomscale/timeseries/physical_sample.py",
|
|
106
107
|
"src/atomscale/timeseries/polling.py",
|
|
107
108
|
"src/atomscale/timeseries/provider.py",
|
|
108
109
|
"src/atomscale/timeseries/recipe.py",
|
|
@@ -124,6 +125,7 @@
|
|
|
124
125
|
"tests/test_metrology.py",
|
|
125
126
|
"tests/test_optical.py",
|
|
126
127
|
"tests/test_photoluminescence.py",
|
|
128
|
+
"tests/test_physical_sample_timeseries.py",
|
|
127
129
|
"tests/test_polling.py",
|
|
128
130
|
"tests/test_raman.py",
|
|
129
131
|
"tests/test_recipe.py",
|