Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,111 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
license: mit
|
| 5 |
+
task_categories:
|
| 6 |
+
- text-classification
|
| 7 |
+
tags:
|
| 8 |
+
- clarus
|
| 9 |
+
- f1
|
| 10 |
+
- thermal
|
| 11 |
+
- latent-instability
|
| 12 |
+
- cross-coupling
|
| 13 |
+
size_categories:
|
| 14 |
+
- 1K<n<10K
|
| 15 |
+
pretty_name: F1 Latent Cross-Coupling Thermal Load Instability v0.1
|
| 16 |
+
---
|
| 17 |
+
|
| 18 |
+
# What this repo does
|
| 19 |
+
|
| 20 |
+
Detects hidden thermal instability before performance loss appears.
|
| 21 |
+
|
| 22 |
+
Focus: temperature-driven failure across interacting systems.
|
| 23 |
+
|
| 24 |
+
# Core variables
|
| 25 |
+
|
| 26 |
+
- tyre_temp_load
|
| 27 |
+
- brake_temp_load
|
| 28 |
+
- power_unit_heat_load
|
| 29 |
+
- cooling_efficiency
|
| 30 |
+
|
| 31 |
+
# Prediction target
|
| 32 |
+
|
| 33 |
+
`label_thermal_load_instability`
|
| 34 |
+
|
| 35 |
+
1 → thermal regime will force performance drop
|
| 36 |
+
0 → thermal state remains stable
|
| 37 |
+
|
| 38 |
+
# Key idea
|
| 39 |
+
|
| 40 |
+
Thermal failure is rarely single-source.
|
| 41 |
+
|
| 42 |
+
It emerges from interaction:
|
| 43 |
+
|
| 44 |
+
- tyre overheating
|
| 45 |
+
- brake heat transfer
|
| 46 |
+
- power unit load
|
| 47 |
+
- cooling limits
|
| 48 |
+
|
| 49 |
+
# Label logic
|
| 50 |
+
|
| 51 |
+
label = 1 if:
|
| 52 |
+
|
| 53 |
+
- latent_instability_score ≥ 0.60
|
| 54 |
+
- cross_coupling_intensity ≥ 0.60
|
| 55 |
+
- hidden_state_index ≥ 0.60
|
| 56 |
+
- activation_threshold_distance ≤ 0.35
|
| 57 |
+
- tyre_temp_load ≥ 0.80
|
| 58 |
+
- brake_temp_load ≥ 0.78
|
| 59 |
+
- power_unit_heat_load ≥ 0.75
|
| 60 |
+
- tyre_temp_load > stabilization_buffer
|
| 61 |
+
|
| 62 |
+
# Why this matters
|
| 63 |
+
|
| 64 |
+
Performance loss is often thermal before it is visible.
|
| 65 |
+
|
| 66 |
+
This dataset detects:
|
| 67 |
+
|
| 68 |
+
- overheating before lap time drop
|
| 69 |
+
- cooling saturation before failure
|
| 70 |
+
- system interaction before degradation
|
| 71 |
+
|
| 72 |
+
# Use cases
|
| 73 |
+
|
| 74 |
+
- race strategy
|
| 75 |
+
- cooling configuration
|
| 76 |
+
- stint planning
|
| 77 |
+
- simulation
|
| 78 |
+
|
| 79 |
+
# Evaluation
|
| 80 |
+
|
| 81 |
+
Primary:
|
| 82 |
+
missed_latent_activation_rate
|
| 83 |
+
|
| 84 |
+
Secondary:
|
| 85 |
+
false_activation_rate
|
| 86 |
+
|
| 87 |
+
# Structural Note
|
| 88 |
+
|
| 89 |
+
This sits in the latent detection layer.
|
| 90 |
+
|
| 91 |
+
It detects hidden thermal instability before performance collapse appears.
|
| 92 |
+
|
| 93 |
+
# Production Deployment
|
| 94 |
+
|
| 95 |
+
Used for:
|
| 96 |
+
|
| 97 |
+
- live telemetry augmentation
|
| 98 |
+
- predictive race control
|
| 99 |
+
- simulation pipelines
|
| 100 |
+
|
| 101 |
+
# Collaboration
|
| 102 |
+
|
| 103 |
+
Suitable for:
|
| 104 |
+
|
| 105 |
+
- F1 teams
|
| 106 |
+
- motorsport analytics
|
| 107 |
+
- simulation groups
|
| 108 |
+
|
| 109 |
+
# License
|
| 110 |
+
|
| 111 |
+
MIT
|