Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,84 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
license: mit
|
| 5 |
+
pretty_name: EUV Projection Optics Deformation–Overlay Coherence Drift Detection v0.1
|
| 6 |
+
dataset_name: euv-projection-optics-deformation-overlay-coherence-drift-detection-v0.1
|
| 7 |
+
tags:
|
| 8 |
+
- clarusc64
|
| 9 |
+
- euv
|
| 10 |
+
- lithography
|
| 11 |
+
- projection-optics
|
| 12 |
+
- overlay
|
| 13 |
+
- drift-detection
|
| 14 |
+
task_categories:
|
| 15 |
+
- tabular-classification
|
| 16 |
+
- tabular-regression
|
| 17 |
+
size_categories:
|
| 18 |
+
- 1K<n<10K
|
| 19 |
+
configs:
|
| 20 |
+
- config_name: default
|
| 21 |
+
data_files:
|
| 22 |
+
- split: train
|
| 23 |
+
path: data/train.csv
|
| 24 |
+
- split: test
|
| 25 |
+
path: data/test.csv
|
| 26 |
+
---
|
| 27 |
+
|
| 28 |
+
## What this dataset tests
|
| 29 |
+
|
| 30 |
+
When thermo-mechanical deformation stops predicting overlay error cleanly.
|
| 31 |
+
|
| 32 |
+
This is the second layer of projection optics thermo-mechanical decoupling:
|
| 33 |
+
|
| 34 |
+
deformation modes
|
| 35 |
+
overlay error growth
|
| 36 |
+
cause-axis attribution
|
| 37 |
+
|
| 38 |
+
## Task
|
| 39 |
+
|
| 40 |
+
Given deformation modes and overlay state, predict:
|
| 41 |
+
|
| 42 |
+
overlay_drift_score (0..1)
|
| 43 |
+
drift_flag (0/1)
|
| 44 |
+
dominant_cause_axis
|
| 45 |
+
|
| 46 |
+
Output format:
|
| 47 |
+
float,int,string
|
| 48 |
+
|
| 49 |
+
Example:
|
| 50 |
+
0.58,1,cooling
|
| 51 |
+
|
| 52 |
+
## Fields
|
| 53 |
+
|
| 54 |
+
Inputs
|
| 55 |
+
- thermal_mode1_nm
|
| 56 |
+
- thermal_mode2_nm
|
| 57 |
+
- thermal_mode3_nm
|
| 58 |
+
- overlay_error_nm
|
| 59 |
+
- overlay_within_spec
|
| 60 |
+
- baseline_coherence
|
| 61 |
+
|
| 62 |
+
Targets
|
| 63 |
+
- overlay_drift_score
|
| 64 |
+
- drift_flag
|
| 65 |
+
- dominant_cause_axis
|
| 66 |
+
|
| 67 |
+
## Cause axis meanings
|
| 68 |
+
|
| 69 |
+
thermal: global heating trend
|
| 70 |
+
cooling: flow imbalance or coolant loss
|
| 71 |
+
mount: clamp or support drift
|
| 72 |
+
metrology: measurement/model mismatch
|
| 73 |
+
contamination: absorbed power rise from fouling
|
| 74 |
+
none: stable regime
|
| 75 |
+
|
| 76 |
+
## Evaluation
|
| 77 |
+
|
| 78 |
+
Absolute error for drift score
|
| 79 |
+
Bonus for correct drift flag
|
| 80 |
+
Bonus for correct cause axis
|
| 81 |
+
|
| 82 |
+
## Version
|
| 83 |
+
|
| 84 |
+
v0.1
|