VLM Data Doctor
Offline · Read-only · CPUImage-text SFT / Data quality report
Dataset preflight.
No issues found by this profile. Preserve this report and validate the training pipeline next.
Source records3train 2 · eval 1
Errors0Findings that block this profile
Warnings0Review against your task policy
Decoded image files3Unique resolved paths
Findings by rule
No findings from the supported checks.
Counts are findings, not distinct failed records. Showing up to six rules.
Your next steps
1
Review the affected rows
Resolve errors and decide whether warnings violate your split policy.
2
Re-run and preserve the evidence
Archive the JSON report, input hashes and experiment configuration.
3
Run a small training smoke test
Template, tokenizer and model behavior still need separate verification.
Review queue
0 findings| Severity | Rule | Location | What to do |
|---|
No findings match this view.
Reproducibility manifest · input hashes and settings
{
"tool": "vlm-data-doctor",
"version": "0.3.1",
"profile": "image-text-sft-v2",
"python": "3.12.15",
"pillow": "12.3.0",
"settings": {
"max_pixels": 40000000,
"group_key": null
},
"dataset_sha256": {
"train": "014d4499ffe92f0a3189cd346db170a2a6f6fc58a622ba64eac90e3b1470b383",
"eval": "d0d81127a9d592e0284de923623e4816f4d62cd2270121109367fdd887126e9f"
},
"decoded_image_inventory_sha256": "53ef9770acd7bf53ec5f43b48df947ac05db3374875bdb4f21b9dc3fc3656345"
}