PHYSICALTEST APPLY RESULT TEMPLATE WORKER¶
Applies a response template to one or more physical tests to extract and optionally persist structured responses (KPIs). Use this worker to batch-process physical test records against a defined response template, with control over pagination, re-extraction, and auto-save behaviour.
When to use¶
Tagged: apply_template, auto_save, batch, extraction, pagination, physical_test, reextract, response_template.
Inputs¶
| Label | ID | Type | Default | Required | Description |
|---|---|---|---|---|---|
| Physicaltest | apply_template_worker_pt_id | remote_lookup | — | ✓ | One or more physical test records (remote lookup into the physicaltests datatype) to which the response template will be applied; required. |
| Response Template | apply_template_worker_rt_id | remote_lookup | — | The response template (remote lookup into the templates datatype) that defines which channels/KPIs to extract; leave blank to use the template already associated with each physical test. | |
| Save Extracted Responses | apply_template_worker_rt_auto_save | select | no | ✓ | Controls whether extracted responses are written back to the physical test record (yes) or returned transiently without saving (no, default); choose no during development/debugging to avoid overwriting existing responses. |
| Re-extract Existing Responses | reextract | select | no | ✓ | When set to yes, forces a full re-extraction of channels even if responses already exist for the test; default no re-uses previously extracted responses to save processing time. |
| Number of responses per page | num_responses_per_page | text | -1 | Maximum number of responses returned per page; default -1 returns all responses in a single page; set a positive integer together with page_num to paginate large result sets. | |
| Page Number | page_num | text | 1 | 1-based page index to retrieve when paginating results; default 1 (first page); only meaningful when num_responses_per_page is a positive integer. | |
| Max File Size | max_value_size | text | 0 | Target maximum payload size (bytes) used to auto-select the number of records per page; default 0 disables size-based pagination and defers to num_responses_per_page. | |
| Save Dataset to File | save_to_file | select | no | When yes, write the consolidated responses as a tests-as-rows dataset FILE. | |
| Dataset File Format | file_format | select | csv | Format of the saved dataset file when save_to_file=yes. csv/tsv digitize each curve into digitize_points columns and stream with bounded memory. (JSON was removed – it is a single non-streamable document that must be fully decoded in PHP, defeating the memory-bounded file path and causing OOM on large datasets; use csv/tsv.) | |
| Curve Digitize Points | digitize_points | text | 100 | Number of points each curve response is digitized to when writing csv/tsv columns (ignored for json). Default 100. | |
| Dataset Name | dataset_name | text | — | Optional name for the Dataset record created when save_to_file=yes. Defaults to ‘Physical Test Result Template Dataset <date>’. | |
| Target Response(s) | target_responses | text | — | Comma-separated response name(s) to treat as the ML TARGET (e.g. “SOC”). Used only when save_to_file=yes to split the dataset columns into the independents_csv (features) and clabel_csv (targets) outputs you can wire straight into ml_learn_auto. | |
| Include Data Sources | include_data_sources | text | — | Comma-separated data source(s) to KEEP (e.g. “binout,d3plot”). When non-empty, only responses whose data_source matches are included; everything else is dropped. Leave blank to include all. Case-insensitive. | |
| Skip Data Sources | skip_data_sources | text | — | Comma-separated data source(s) to DROP (e.g. “manual” to exclude manually-authored scalar responses like CLI-args strings). Applied after Include Data Sources. Leave blank to skip none. Case-insensitive. | |
| Skip Errored Responses | skip_errored_responses | select | no | When yes, a single response that errors while its output is built becomes a warning + skip instead of failing the whole worker. | |
| Exclude Per-Response Outputs | exclude_responses_from_output | select | no | When yes, the worker does NOT emit one response_<id> output per response. For a large template across many tests this prevents tens of thousands of individual outputs. The consolidated responses, responses_by_test, and dataset outputs (plus dataset_file) still carry every response, so nothing is lost for ML/batch use. |
Outputs¶
| Label | ID | Type | Description |
|---|---|---|---|
| Physicaltest | physicaltest_id | dataset | Dataset of the physical test record(s) that were processed, returned as a physicaltests dataType for downstream reference or chaining. |
| Dataset File | dataset_file | file | Vault path of the tests-as-rows dataset file (only when save_to_file=yes). Pass to a Lucy ML training worker (ml_learn_auto dataset_file input) to train directly from the file. |
| Dataset ID | dataset_id | integer | ID of the Dataset record created for the saved dataset file (only when save_to_file=yes). |
| Dataset Columns | dataset_columns | json | Ordered header of the saved dataset file (only when save_to_file=yes). |
| Feature Columns (CSV) | independents_csv | string | Comma-separated feature column names (all data columns except the target). Wire into ml_learn_auto independents_csv. Only when save_to_file=yes. |
| Target Columns (CSV) | clabel_csv | string | Comma-separated target column names (the columns of target_responses). Wire into ml_learn_auto clabel_csv. Only when save_to_file=yes. |
| Responses | responses | dataset | Dataset of extracted response objects (responses dataType) containing the channel values and KPIs produced by applying the template; reflects the current page when pagination is active. |
| Responses by Test | responses_by_test | dataset | Pivoted dataset with one row per response name and one column per physical test (column header is the test name); each cell holds that response’s value for that test. Useful for comparing the same KPI across multiple selected tests. |
| Dataset (tests as rows) | dataset | dataset | The tests-as-rows ML matrix in memory (one row per test; each curve digitized to digitize_points columns). Feed a batch straight into cleaning/normalize/PCA/ml workers without a file hop. Bounded by the number of tests, not curve sizes. |
| Page | page | integer | Integer indicating the page number of the response dataset that was returned; useful for loop-based pagination in a parent workflow. |
| Records per Page | limit | integer | Integer indicating the actual number of records per page that was used for this execution, which may be auto-computed from max_value_size if that parameter is non-zero. |
Disciplines¶
- cae.postprocessing.response
- data.dataset.ingest
- platform.job_submission
Auto-generated from platform schema. Worker id: physicaltest_apply_result_template. Schema hash: 96f0a620cd8e. Hand-curated docs in workerexamples/ override this page when present.