What this category's data looks like
In the course scheduling and course consumption scenario of training institutions, materials come from class schedules, course consumption records, schedule change requests, teacher sign-in sheets, and receipt forms. The update rhythm is weekly scheduling, per-session course consumption, and on-demand schedule changes. Fields include course name, class, teacher, classroom, class date and time slot, number of class hours, student name, signature, and receipt status. During multilingual materials and terminology processing, the class schedule serves as the baseline for courses and time slots, course consumption records serve as the actual consumption and signature evidence, and schedule change requests serve as the before-and-after comparison of changes. After these three enter the document parsing and chunking stage, each forms retrievable data rows, and the terminology glossary and comparison draft are aggregated and generated during the retrieval and subsequent processing stage.
Configuration and implementation checks
This step is executed by parsing and training tasks (server-side tasks), outputting data rows and index statuses (parsed, indexing, indexed). The unified terminology glossary and comparison draft are then completed by retrieval and subsequent processing.
| Object | Check | Verification action |
|---|---|---|
| Class schedule | Course name, class, teacher, classroom, class date and time slot, and number of class hours are written consistently across all materials | Open the data rows to check the status, separately verify the file records and data rows, and confirm that the translated names of class schedule terminology throughout the document can refer back to the source text |
| Course consumption record | The student, number of class hours, date, and signature of the same course consumption record match across the two materials | Sample three terms to check whether their written forms are consistent throughout the document, verify whether signatures and receipts are complete, and register the number of array entries and the tokens of the referenced text separately |
| Schedule change request | The course, date, teacher, and classroom fields before and after the schedule change correspond to the class schedule and course consumption records | Verify the fields, dates, and signatures of the schedule change request against the class schedule and course consumption records, and confirm that versions are synchronized and no receipts are missing |
| Parsing and training tasks (server-side tasks) | Data rows are in one of the three states parsed, indexing, or indexed and are consistent with the interface display | Verify the data rows corresponding to chunkSize, chunkSplitter, and trainingType, and check whether the data rows covered by the rebuild task include only data that has completed indexing |
Three common errors
- Symptom: The course name, number of class hours, and date of the same course consumption record do not match between the class schedule and the course consumption record. Cause: The versions of the two materials are not synchronized, and the translated terminology names each follow old drafts.
- Symptom: The fields of the schedule change request are inconsistent with the class schedule and course consumption records, and signatures or receipts are missing. Cause: Changes were not written back throughout the document, and signatures and receipts were not updated with the version.
- Symptom: Newly uploaded materials cannot be found, and empty data rows remain after training fails. Cause: Successful parsing does not equal retrieval hits, the index status has not reached completed, or the rebuild scope and the initial training coverage scope have been swapped.
How to verify the setup
Acceptance for this feature: Upload one material, observe the status transitions of data rows and the coverage scope of the rebuild task, verify that the three states parsed, indexing, and indexed are consistent with the interface display, and when changing models, rebuild only data that has completed indexing.
Business test: Sample three terms to check whether their written forms are consistent throughout the document, confirm that the translated names can refer back to the source text, and separately record the number of array entries and the tokens of the referenced text.
- Open the data rows to check the status, separately verify the file records and data rows, and confirm that the identifiers of files and data rows can be distinguished.
- Verify the data rows corresponding to
chunkSize,chunkSplitter, andtrainingType, and check whether the data rows covered by the rebuild task include only data that has completed indexing.