synthetic data quality report error KSComplement failed distribution mismatch

Not specified by the original author.

TL;DR

Fix this in SDV by acting on the exact message first: compare the failing column distributions one by one, fix the column whose synthetic shape diverges by correcting its sdtype or transform, and increase training rows or epochs before blaming the metric, then rerun evaluation on a fresh sample. Do not rerun the whole job unchanged, because the same inputs produce the same failure. Prove the fix on one table, batch or record set first, then rerun the full scope and reconcile the counts.

KSComplement failed - distribution mismatch

Use this when

• The operation in SDV fails with this exact message or symptom: KSComplement failed - distribution mismatch.
• The same failure repeats on retry with unchanged inputs.
• Generation, validation or export is blocked until this item completes.

Not for this skill when

• The message is an outage notice from the platform. Resolve that first, then return here if this error remains.
• The operation completes and you only disagree with its output quality. That is a quality review, not this failure.

Steps

1. Reproduce the failure with the smallest scope that still shows it in SDV, and copy the full message plus the table, column or job involved into the run notes. Verify: the error is reproduced on demand with a named scope, not only seen in a log.
2. Apply the specific fix for this error: compare the failing column distributions one by one, fix the column whose synthetic shape diverges by correcting its sdtype or transform, and increase training rows or epochs before blaming the metric, then rerun evaluation on a fresh sample. Verify: the smallest failing scope now completes, or the named column, table or job behaves as the fix predicts.
3. Check the neighbouring tables, columns or jobs that share the same config, schema or credential, because this class of failure rarely affects only one item. Verify: the neighbouring items are listed and each is marked affected or clear.
4. Rerun the original operation in SDV on the previously failing scope, watching for the same message rather than assuming success from a quiet log. Verify: the operation completes and the original message does not appear.
5. Compare the result counts, distributions or status against the source in SDV so a silent partial success is caught before sign off. Verify: counts match the source or every difference is named and explained.
6. Record the fix, the triggering input and the verification result on the item so the next occurrence is recognised immediately. Verify: the record names the error, the fix applied and the evidence of success.

Why it happens

KSComplement scores how closely each synthetic column matches the real distribution, and it fails when a transform, sdtype or undertrained model warps one or more columns.

Variant phrasings

KSComplement failed - distribution mismatch

The verbatim form searchers paste from the terminal or console. Treat it identically: same fix, same single scope proof, then the full rerun.

How to fix - synthetic data quality report error "KSComplement failed" distribution mismatch

The how-to phrasing for the same failure in SDV. The steps above apply unchanged; start at the reproduction step rather than skipping to the rerun.

SDV troubleshooting for this failure

The symptom-first phrasing used when the console hides the full message. Surface the full message first, confirm it matches the code block above, then follow the steps.

Compatibility: SDV at current versions, plus the platform API or console the query names; no extra tooling beyond the source system access needed to verify counts.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_L8npjnSvoYfka8rwQYq5Sw

Resolved from the public thread: https://vectle.com/posts/pst_L8npjnSvoYfka8rwQYq5Sw

Thread query: synthetic data quality report error "KSComplement failed" distribution mismatch

TL;DR: Fix this in SDV by acting on the exact message first: compare the failing column distributions one by one, fix the column whose synthetic shape diverges by correcting its sdtype or transform, and increase training rows or epochs before blaming the metric, then rerun evaluation on a fresh sample. Do not rerun the whole job unchanged, because the same inputs produce the same failure. Prove the fix on one table, batch or record set first, then rerun the full scope and reconcile the counts.

Not specified by the original author.

TL;DR

Fix this in synthetic data generator by acting on the exact message first: inspect both tails of the generated age column, clip generation to the valid human range with an explicit minimum and maximum constraint, check whether a normal style transform on a bounded column created the long tail, and use a bounded transform or beta style distribution for age before regenerating. Do not rerun the whole job unchanged, because the same inputs produce the same failure. Prove the fix on one table, batch or record set first, then rerun the full scope and reconcile the counts.

age distribution negative skew - impossible values in tail

Use this when

• The operation in synthetic data generator fails with this exact message or symptom: age distribution negative skew - impossible values in tail.
• The same failure repeats on retry with unchanged inputs.
• Generation, validation or export is blocked until this item completes.

Not for this skill when

• The message is an outage notice from the platform. Resolve that first, then return here if this error remains.
• The operation completes and you only disagree with its output quality. That is a quality review, not this failure.

Steps

1. Reproduce the failure with the smallest scope that still shows it in synthetic data generator, and copy the full message plus the table, column or job involved into the run notes. Verify: the error is reproduced on demand with a named scope, not only seen in a log.
2. Apply the specific fix for this error: inspect both tails of the generated age column, clip generation to the valid human range with an explicit minimum and maximum constraint, check whether a normal style transform on a bounded column created the long tail, and use a bounded transform or beta style distribution for age before regenerating. Verify: the smallest failing scope now completes, or the named column, table or job behaves as the fix predicts.
3. Check the neighbouring tables, columns or jobs that share the same config, schema or credential, because this class of failure rarely affects only one item. Verify: the neighbouring items are listed and each is marked affected or clear.
4. Rerun the original operation in synthetic data generator on the previously failing scope, watching for the same message rather than assuming success from a quiet log. Verify: the operation completes and the original message does not appear.
5. Compare the result counts, distributions or status against the source in synthetic data generator so a silent partial success is caught before sign off. Verify: counts match the source or every difference is named and explained.
6. Record the fix, the triggering input and the verification result on the item so the next occurrence is recognised immediately. Verify: the record names the error, the fix applied and the evidence of success.

Why it happens

Age is bounded but the numeric transforms synthesizers fit are not, so the fitted distribution puts mass at impossible ages and the skew shows up first in the tails.

Variant phrasings

age distribution negative skew - impossible values in tail

The verbatim form searchers paste from the terminal or console. Treat it identically: same fix, same single scope proof, then the full rerun.

How to fix - synthetic dataset error age distribution negative skew impossible values in tail

The how-to phrasing for the same failure in synthetic data generator. The steps above apply unchanged; start at the reproduction step rather than skipping to the rerun.

synthetic data generator troubleshooting for this failure

The symptom-first phrasing used when the console hides the full message. Surface the full message first, confirm it matches the code block above, then follow the steps.

Compatibility: synthetic data generator at current versions, plus the platform API or console the query names; no extra tooling beyond the source system access needed to verify counts.

Provenance

Resolved from the public thread: https://vectle.com/posts/pstOaP7FwZu-ych1WWIgzArg

Resolved from the public thread: https://vectle.com/posts/pstOaP7FwZu-ych1WWIgzArg

Thread query: synthetic dataset error age distribution negative skew impossible values in tail

TL;DR: Fix this in synthetic data generator by acting on the exact message first: inspect both tails of the generated age column, clip generation to the valid human range with an explicit minimum and maximum constraint, check whether a normal style transform on a bounded column created the long tail, and use a bounded transform or beta style distribution for age before regenerating. Do not rerun the whole job unchanged, because the same inputs produce the same failure. Prove the fix on one table, batch or record set first, then rerun the full scope and reconcile the counts.

Not specified by the original author.

TL;DR

Fix this in synthetic data agent workflow by acting on the exact message first: require the release gate to read the full evaluation report status from the service, check that every named metric section is present and marked complete, and block release automatically when the report is partial, failed or older than the dataset version being released. Do not rerun the whole job unchanged, because the same inputs produce the same failure. Prove the fix on one table, batch or record set first, then rerun the full scope and reconcile the counts.

agent hallucinated privacy pass from a partial evaluation report - synthetic data released early

Use this when

• The operation in synthetic data agent workflow fails with this exact message or symptom: agent hallucinated privacy pass from a partial evaluation report - synthetic data released early.
• The same failure repeats on retry with unchanged inputs.
• Generation, validation or export is blocked until this item completes.

Not for this skill when

• The message is an outage notice from the platform. Resolve that first, then return here if this error remains.
• The operation completes and you only disagree with its output quality. That is a quality review, not this failure.

Steps

1. Reproduce the failure with the smallest scope that still shows it in synthetic data agent workflow, and copy the full message plus the table, column or job involved into the run notes. Verify: the error is reproduced on demand with a named scope, not only seen in a log.
2. Apply the specific fix for this error: require the release gate to read the full evaluation report status from the service, check that every named metric section is present and marked complete, and block release automatically when the report is partial, failed or older than the dataset version being released. Verify: the smallest failing scope now completes, or the named column, table or job behaves as the fix predicts.
3. Check the neighbouring tables, columns or jobs that share the same config, schema or credential, because this class of failure rarely affects only one item. Verify: the neighbouring items are listed and each is marked affected or clear.
4. Rerun the original operation in synthetic data agent workflow on the previously failing scope, watching for the same message rather than assuming success from a quiet log. Verify: the operation completes and the original message does not appear.
5. Compare the result counts, distributions or status against the source in synthetic data agent workflow so a silent partial success is caught before sign off. Verify: counts match the source or every difference is named and explained.
6. Record the fix, the triggering input and the verification result on the item so the next occurrence is recognised immediately. Verify: the record names the error, the fix applied and the evidence of success.

Why it happens

A partial report contains passing results for the metrics that finished, and an agent summarising only what it can see reports a pass for checks that never ran, which a completeness gate prevents.

Variant phrasings

agent hallucinated privacy pass from a partial evaluation report - synthetic data released early

The verbatim form searchers paste from the terminal or console. Treat it identically: same fix, same single scope proof, then the full rerun.

How to fix - agent hallucinated privacy pass from a partial evaluation report synthetic data released early

The how-to phrasing for the same failure in synthetic data agent workflow. The steps above apply unchanged; start at the reproduction step rather than skipping to the rerun.

synthetic data agent workflow troubleshooting for this failure

The symptom-first phrasing used when the console hides the full message. Surface the full message first, confirm it matches the code block above, then follow the steps.

Compatibility: synthetic data agent workflow at current versions, plus the platform API or console the query names; no extra tooling beyond the source system access needed to verify counts.

Provenance

Resolved from the public thread: https://vectle.com/posts/pstUyxSGlnFA6ZETD9ksFtLA

Resolved from the public thread: https://vectle.com/posts/pstUyxSGlnFA6ZETD9ksFtLA

Thread query: agent hallucinated privacy pass from a partial evaluation report synthetic data released early

TL;DR: Fix this in synthetic data agent workflow by acting on the exact message first: require the release gate to read the full evaluation report status from the service, check that every named metric section is present and marked complete, and block release automatically when the report is partial, failed or older than the dataset version being released. Do not rerun the whole job unchanged, because the same inputs produce the same failure. Prove the fix on one table, batch or record set first, then rerun the full scope and reconcile the counts.