Dataset Viewer
Auto-converted to Parquet Duplicate
scenario_id
string
pressure
float64
buffer_capacity
float64
drift_gradient
float64
boundary_distance
float64
recovery_feasibility
float64
label_buffer_exhaustion
int64
be_train_001
0.82
0.14
0.61
0.19
0.21
1
be_train_002
0.77
0.18
0.58
0.23
0.27
1
be_train_003
0.71
0.16
0.54
0.28
0.25
1
be_train_004
0.88
0.12
0.67
0.17
0.19
1
be_train_005
0.74
0.21
0.49
0.31
0.34
1
be_train_006
0.39
0.66
-0.22
0.71
0.68
0
be_train_007
0.33
0.73
-0.27
0.76
0.72
0
be_train_008
0.48
0.59
-0.11
0.63
0.61
0
be_train_009
0.41
0.69
-0.18
0.68
0.66
0
be_train_010
0.52
0.55
-0.09
0.57
0.59
0

What this dataset tests

This dataset tests whether a model can recognize instability caused by buffer exhaustion from a compact state-space snapshot.

Core instability geometry

Buffer exhaustion occurs when system load rises while protective capacity is steadily depleted. Collapse occurs when remaining buffers are insufficient to absorb further strain.

Prediction target

label_buffer_exhaustion

1 = instability due to buffer exhaustion
0 = system remains stable

Row structure

Each row represents a system state.

Columns

scenario_id
pressure
buffer_capacity
drift_gradient
boundary_distance
recovery_feasibility
label_buffer_exhaustion (train only)

Signal definitions

pressure
Current system load or strain level.

buffer_capacity
Remaining capacity available to absorb strain.

drift_gradient
Direction and strength of movement toward instability.

boundary_distance
Distance from the instability boundary.

recovery_feasibility
Likelihood the system can return to stability.

Files

data/train.csv
10 labeled training rows

data/tester.csv
10 unlabeled evaluation rows

scorer.py
binary classification scorer

README.md
dataset card

Evaluation

Prediction file format

scenario_id,prediction

Example

be_test_001,1
be_test_002,0

Run scorer

python scorer.py predictions.csv ground_truth.csv

Why this matters

Many complex systems collapse because protective buffers are depleted.
Recognizing this geometry is central to stability intelligence.

License

MIT

Downloads last month
21