TunnelInstruct / README.md
augustusper's picture
Upload 47 files
6a3b2fe verified
|
Raw History Blame Contribute Delete
3.72 kB
metadata
language:
  - en
task_categories:
  - image-text-to-text
tags:
  - tunnel-inspection
  - infrastructure
  - TunnelWeave
size_categories:
  - 1K<n<10K
configs:
  - config_name: all
    default: true
    data_files:
      - split: train
        path: data--*--train-*.parquet
  - config_name: crack
    data_files:
      - split: train
        path: data--crack--train-*.parquet
  - config_name: deformed
    data_files:
      - split: train
        path: data--deformed--train-*.parquet
  - config_name: detached
    data_files:
      - split: train
        path: data--detached--train-*.parquet
  - config_name: displacement
    data_files:
      - split: train
        path: data--displacement--train-*.parquet
  - config_name: exposed_rebar
    data_files:
      - split: train
        path: data--exposed_rebar--train-*.parquet
  - config_name: foreign_object
    data_files:
      - split: train
        path: data--foreign_object--train-*.parquet
  - config_name: fracture
    data_files:
      - split: train
        path: data--fracture--train-*.parquet
  - config_name: inverted
    data_files:
      - split: train
        path: data--inverted--train-*.parquet
  - config_name: missing
    data_files:
      - split: train
        path: data--missing--train-*.parquet
  - config_name: normal
    data_files:
      - split: train
        path: data--normal--train-*.parquet
  - config_name: rust
    data_files:
      - split: train
        path: data--rust--train-*.parquet
  - config_name: spalling
    data_files:
      - split: train
        path: data--spalling--train-*.parquet
  - config_name: surface_deposit
    data_files:
      - split: train
        path: data--surface_deposit--train-*.parquet
  - config_name: surface_wetness
    data_files:
      - split: train
        path: data--surface_wetness--train-*.parquet

TunnelInstruct

1,740 original instruction-response records over 1,725 unique training images. The original class-conditioned training instruction and response are reconstructed with the frozen historical event_task function; context/evidence/assessment are copied verbatim. Special output tokens belong to that training interface. Descriptions are model-assisted references, not independently human-certified diagnoses. This is the historical TunnelInstruct collection, distinct from the later 192-reference image-facts ablation subset.

Organization

Native images and instruction-response records are embedded in Parquet shards organized by the target category. samples.jsonl identifies each record and index--*.jsonl provides the category indices. The fourteen configurations comprise thirteen defect categories and normal examples; all loads the complete collection. Multiple records may use the same image with different target categories.

from datasets import load_dataset
ds = load_dataset("augustusper/TunnelInstruct", "all", split="train", token=True)

Source and usage

Source datasets: TACK (https://arxiv.org/abs/2512.14477), dacl10k (https://openaccess.thecvf.com/content/WACV2024/html/Flotzinger_dacl10k_Benchmark_for_Semantic_Bridge_Damage_Segmentation_WACV_2024_paper.html), RFDD (https://doi.org/10.1038/s41597-026-07851-7), the CTCD release used by the project, and author-collected tunnel imagery. See per-record source_dataset. Original source terms continue to apply. This private research archive does not establish a blanket redistribution license; public release requires source and author permissions. No test, validation, or cross-tunnel diagnostic split is included.

Source and synthetic provenance are retained per record. Images are not resized or regenerated. No private file-system paths, credentials, run logs, or model weights are included. export_summary.json and SHA256SUMS record the package contents.