{ "about": "Golden values from plumbline.metrics.calibration (synthetic_floor for the calculator, ece_figure for pasted rows), which site/floor.js must reproduce. Written by scripts/floor_golden.py.", "function": "synthetic_floor(n, n_bins, accuracy), all other arguments default", "defaults": { "binning": "equal_width", "n_boot": 2000, "concentration": 6.0, "seed": 0 }, "numpy": "2.5.3", "tolerance": 1e-09, "pcg64_initial_state": { "0": { "state": "35399562948360463058890781895381311971", "inc": "87136372517582989555478159403783844777" }, "1": { "state": "207833532711051698738587646355624148094", "inc": "194290289479364712180083596243593368443" } }, "cases": [ { "n": 1, "n_bins": 10, "accuracy": 0.8, "mean": 0.3800222137274552, "p95": 0.7463609574384902, "statements": { "0.00125": "ECE 0.0013 over 1 rows (10 equal width bins), against a calibrated-model floor of 0.3800 (95th percentile 0.7464): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 1 rows (10 equal width bins), against a calibrated-model floor of 0.3800 (95th percentile 0.7464): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 1 rows (10 equal width bins), against a calibrated-model floor of 0.3800 (95th percentile 0.7464): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 1 rows (10 equal width bins), against a calibrated-model floor of 0.3800 (95th percentile 0.7464): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 1 rows (10 equal width bins), against a calibrated-model floor of 0.3800 (95th percentile 0.7464): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 1 rows (10 equal width bins), against a calibrated-model floor of 0.3800 (95th percentile 0.7464): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 1 rows (10 equal width bins), against a calibrated-model floor of 0.3800 (95th percentile 0.7464): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 1 rows (10 equal width bins), against a calibrated-model floor of 0.3800 (95th percentile 0.7464): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable." } }, { "n": 5, "n_bins": 10, "accuracy": 0.8, "mean": 0.2847577610479967, "p95": 0.42849582212750476, "statements": { "0.00125": "ECE 0.0013 over 5 rows (10 equal width bins), against a calibrated-model floor of 0.2848 (95th percentile 0.4285): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 5 rows (10 equal width bins), against a calibrated-model floor of 0.2848 (95th percentile 0.4285): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 5 rows (10 equal width bins), against a calibrated-model floor of 0.2848 (95th percentile 0.4285): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 5 rows (10 equal width bins), against a calibrated-model floor of 0.2848 (95th percentile 0.4285): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 5 rows (10 equal width bins), against a calibrated-model floor of 0.2848 (95th percentile 0.4285): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 5 rows (10 equal width bins), against a calibrated-model floor of 0.2848 (95th percentile 0.4285): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 5 rows (10 equal width bins), against a calibrated-model floor of 0.2848 (95th percentile 0.4285): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 5 rows (10 equal width bins), against a calibrated-model floor of 0.2848 (95th percentile 0.4285): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable." } }, { "n": 5, "n_bins": 20, "accuracy": 0.5, "mean": 0.4039633559253163, "p95": 0.555938114043049, "statements": { "0.00125": "ECE 0.0013 over 5 rows (20 equal width bins), against a calibrated-model floor of 0.4040 (95th percentile 0.5559): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 5 rows (20 equal width bins), against a calibrated-model floor of 0.4040 (95th percentile 0.5559): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 5 rows (20 equal width bins), against a calibrated-model floor of 0.4040 (95th percentile 0.5559): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 5 rows (20 equal width bins), against a calibrated-model floor of 0.4040 (95th percentile 0.5559): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 5 rows (20 equal width bins), against a calibrated-model floor of 0.4040 (95th percentile 0.5559): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 5 rows (20 equal width bins), against a calibrated-model floor of 0.4040 (95th percentile 0.5559): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 5 rows (20 equal width bins), against a calibrated-model floor of 0.4040 (95th percentile 0.5559): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 5 rows (20 equal width bins), against a calibrated-model floor of 0.4040 (95th percentile 0.5559): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable." } }, { "n": 10, "n_bins": 50, "accuracy": 0.8, "mean": 0.27711625440709153, "p95": 0.39266384333888094, "statements": { "0.00125": "ECE 0.0013 over 10 rows (50 equal width bins), against a calibrated-model floor of 0.2771 (95th percentile 0.3927): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 10 rows (50 equal width bins), against a calibrated-model floor of 0.2771 (95th percentile 0.3927): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 10 rows (50 equal width bins), against a calibrated-model floor of 0.2771 (95th percentile 0.3927): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 10 rows (50 equal width bins), against a calibrated-model floor of 0.2771 (95th percentile 0.3927): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 10 rows (50 equal width bins), against a calibrated-model floor of 0.2771 (95th percentile 0.3927): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 10 rows (50 equal width bins), against a calibrated-model floor of 0.2771 (95th percentile 0.3927): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 10 rows (50 equal width bins), against a calibrated-model floor of 0.2771 (95th percentile 0.3927): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 10 rows (50 equal width bins), against a calibrated-model floor of 0.2771 (95th percentile 0.3927): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable." } }, { "n": 20, "n_bins": 10, "accuracy": 0.9, "mean": 0.11220229222944622, "p95": 0.16500137870960746, "statements": { "0.00125": "ECE 0.0013 over 20 rows (10 equal width bins), against a calibrated-model floor of 0.1122 (95th percentile 0.1650): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 20 rows (10 equal width bins), against a calibrated-model floor of 0.1122 (95th percentile 0.1650): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 20 rows (10 equal width bins), against a calibrated-model floor of 0.1122 (95th percentile 0.1650): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 20 rows (10 equal width bins), against a calibrated-model floor of 0.1122 (95th percentile 0.1650): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 20 rows (10 equal width bins), against a calibrated-model floor of 0.1122 (95th percentile 0.1650): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 20 rows (10 equal width bins), against a calibrated-model floor of 0.1122 (95th percentile 0.1650): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 20 rows (10 equal width bins), against a calibrated-model floor of 0.1122 (95th percentile 0.1650): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 20 rows (10 equal width bins), against a calibrated-model floor of 0.1122 (95th percentile 0.1650): miscalibration is distinguishable from sampling noise." } }, { "n": 30, "n_bins": 5, "accuracy": 0.7, "mean": 0.12139637539943718, "p95": 0.20461675476596455, "statements": { "0.00125": "ECE 0.0013 over 30 rows (5 equal width bins), against a calibrated-model floor of 0.1214 (95th percentile 0.2046): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 30 rows (5 equal width bins), against a calibrated-model floor of 0.1214 (95th percentile 0.2046): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 30 rows (5 equal width bins), against a calibrated-model floor of 0.1214 (95th percentile 0.2046): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 30 rows (5 equal width bins), against a calibrated-model floor of 0.1214 (95th percentile 0.2046): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 30 rows (5 equal width bins), against a calibrated-model floor of 0.1214 (95th percentile 0.2046): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 30 rows (5 equal width bins), against a calibrated-model floor of 0.1214 (95th percentile 0.2046): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 30 rows (5 equal width bins), against a calibrated-model floor of 0.1214 (95th percentile 0.2046): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 30 rows (5 equal width bins), against a calibrated-model floor of 0.1214 (95th percentile 0.2046): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable." } }, { "n": 50, "n_bins": 15, "accuracy": 0.8, "mean": 0.13313113466264428, "p95": 0.18497730704057386, "statements": { "0.00125": "ECE 0.0013 over 50 rows (15 equal width bins), against a calibrated-model floor of 0.1331 (95th percentile 0.1850): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 50 rows (15 equal width bins), against a calibrated-model floor of 0.1331 (95th percentile 0.1850): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 50 rows (15 equal width bins), against a calibrated-model floor of 0.1331 (95th percentile 0.1850): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 50 rows (15 equal width bins), against a calibrated-model floor of 0.1331 (95th percentile 0.1850): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 50 rows (15 equal width bins), against a calibrated-model floor of 0.1331 (95th percentile 0.1850): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 50 rows (15 equal width bins), against a calibrated-model floor of 0.1331 (95th percentile 0.1850): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 50 rows (15 equal width bins), against a calibrated-model floor of 0.1331 (95th percentile 0.1850): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 50 rows (15 equal width bins), against a calibrated-model floor of 0.1331 (95th percentile 0.1850): miscalibration is distinguishable from sampling noise." } }, { "n": 100, "n_bins": 10, "accuracy": 0.5, "mean": 0.10154939772542025, "p95": 0.1506344546031062, "statements": { "0.00125": "ECE 0.0013 over 100 rows (10 equal width bins), against a calibrated-model floor of 0.1015 (95th percentile 0.1506): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 100 rows (10 equal width bins), against a calibrated-model floor of 0.1015 (95th percentile 0.1506): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 100 rows (10 equal width bins), against a calibrated-model floor of 0.1015 (95th percentile 0.1506): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 100 rows (10 equal width bins), against a calibrated-model floor of 0.1015 (95th percentile 0.1506): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 100 rows (10 equal width bins), against a calibrated-model floor of 0.1015 (95th percentile 0.1506): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 100 rows (10 equal width bins), against a calibrated-model floor of 0.1015 (95th percentile 0.1506): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 100 rows (10 equal width bins), against a calibrated-model floor of 0.1015 (95th percentile 0.1506): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 100 rows (10 equal width bins), against a calibrated-model floor of 0.1015 (95th percentile 0.1506): miscalibration is distinguishable from sampling noise." } }, { "n": 105, "n_bins": 10, "accuracy": 0.7714, "mean": 0.07507178247177257, "p95": 0.11790988562838273, "statements": { "0.00125": "ECE 0.0013 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0751 (95th percentile 0.1179): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0751 (95th percentile 0.1179): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0751 (95th percentile 0.1179): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0751 (95th percentile 0.1179): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0751 (95th percentile 0.1179): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0751 (95th percentile 0.1179): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0751 (95th percentile 0.1179): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0751 (95th percentile 0.1179): miscalibration is distinguishable from sampling noise." } }, { "n": 105, "n_bins": 10, "accuracy": 0.8, "mean": 0.07199355208097018, "p95": 0.11196474547538811, "statements": { "0.00125": "ECE 0.0013 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0720 (95th percentile 0.1120): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0720 (95th percentile 0.1120): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0720 (95th percentile 0.1120): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0720 (95th percentile 0.1120): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0720 (95th percentile 0.1120): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0720 (95th percentile 0.1120): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0720 (95th percentile 0.1120): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 105 rows (10 equal width bins), against a calibrated-model floor of 0.0720 (95th percentile 0.1120): miscalibration is distinguishable from sampling noise." } }, { "n": 200, "n_bins": 20, "accuracy": 0.6, "mean": 0.09645611455860527, "p95": 0.1281392861832022, "statements": { "0.00125": "ECE 0.0013 over 200 rows (20 equal width bins), against a calibrated-model floor of 0.0965 (95th percentile 0.1281): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 200 rows (20 equal width bins), against a calibrated-model floor of 0.0965 (95th percentile 0.1281): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 200 rows (20 equal width bins), against a calibrated-model floor of 0.0965 (95th percentile 0.1281): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 200 rows (20 equal width bins), against a calibrated-model floor of 0.0965 (95th percentile 0.1281): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 200 rows (20 equal width bins), against a calibrated-model floor of 0.0965 (95th percentile 0.1281): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 200 rows (20 equal width bins), against a calibrated-model floor of 0.0965 (95th percentile 0.1281): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 200 rows (20 equal width bins), against a calibrated-model floor of 0.0965 (95th percentile 0.1281): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 200 rows (20 equal width bins), against a calibrated-model floor of 0.0965 (95th percentile 0.1281): miscalibration is distinguishable from sampling noise." } }, { "n": 250, "n_bins": 100, "accuracy": 0.8, "mean": 0.14023235095799594, "p95": 0.1626327053033523, "statements": { "0.00125": "ECE 0.0013 over 250 rows (100 equal width bins), against a calibrated-model floor of 0.1402 (95th percentile 0.1626): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 250 rows (100 equal width bins), against a calibrated-model floor of 0.1402 (95th percentile 0.1626): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 250 rows (100 equal width bins), against a calibrated-model floor of 0.1402 (95th percentile 0.1626): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 250 rows (100 equal width bins), against a calibrated-model floor of 0.1402 (95th percentile 0.1626): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 250 rows (100 equal width bins), against a calibrated-model floor of 0.1402 (95th percentile 0.1626): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 250 rows (100 equal width bins), against a calibrated-model floor of 0.1402 (95th percentile 0.1626): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.1": "ECE 0.1000 over 250 rows (100 equal width bins), against a calibrated-model floor of 0.1402 (95th percentile 0.1626): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.2": "ECE 0.2000 over 250 rows (100 equal width bins), against a calibrated-model floor of 0.1402 (95th percentile 0.1626): miscalibration is distinguishable from sampling noise." } }, { "n": 300, "n_bins": 10, "accuracy": 0.95, "mean": 0.023666559618355967, "p95": 0.03662071112951838, "statements": { "0.00125": "ECE 0.0013 over 300 rows (10 equal width bins), against a calibrated-model floor of 0.0237 (95th percentile 0.0366): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 300 rows (10 equal width bins), against a calibrated-model floor of 0.0237 (95th percentile 0.0366): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 300 rows (10 equal width bins), against a calibrated-model floor of 0.0237 (95th percentile 0.0366): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 300 rows (10 equal width bins), against a calibrated-model floor of 0.0237 (95th percentile 0.0366): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 300 rows (10 equal width bins), against a calibrated-model floor of 0.0237 (95th percentile 0.0366): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 300 rows (10 equal width bins), against a calibrated-model floor of 0.0237 (95th percentile 0.0366): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 300 rows (10 equal width bins), against a calibrated-model floor of 0.0237 (95th percentile 0.0366): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 300 rows (10 equal width bins), against a calibrated-model floor of 0.0237 (95th percentile 0.0366): miscalibration is distinguishable from sampling noise." } }, { "n": 500, "n_bins": 10, "accuracy": 0.8, "mean": 0.0355612696325439, "p95": 0.053085240536494605, "statements": { "0.00125": "ECE 0.0013 over 500 rows (10 equal width bins), against a calibrated-model floor of 0.0356 (95th percentile 0.0531): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 500 rows (10 equal width bins), against a calibrated-model floor of 0.0356 (95th percentile 0.0531): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 500 rows (10 equal width bins), against a calibrated-model floor of 0.0356 (95th percentile 0.0531): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 500 rows (10 equal width bins), against a calibrated-model floor of 0.0356 (95th percentile 0.0531): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 500 rows (10 equal width bins), against a calibrated-model floor of 0.0356 (95th percentile 0.0531): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 500 rows (10 equal width bins), against a calibrated-model floor of 0.0356 (95th percentile 0.0531): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 500 rows (10 equal width bins), against a calibrated-model floor of 0.0356 (95th percentile 0.0531): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 500 rows (10 equal width bins), against a calibrated-model floor of 0.0356 (95th percentile 0.0531): miscalibration is distinguishable from sampling noise." } }, { "n": 500, "n_bins": 15, "accuracy": 0.9, "mean": 0.02784884432264174, "p95": 0.040260151802465075, "statements": { "0.00125": "ECE 0.0013 over 500 rows (15 equal width bins), against a calibrated-model floor of 0.0278 (95th percentile 0.0403): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 500 rows (15 equal width bins), against a calibrated-model floor of 0.0278 (95th percentile 0.0403): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 500 rows (15 equal width bins), against a calibrated-model floor of 0.0278 (95th percentile 0.0403): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 500 rows (15 equal width bins), against a calibrated-model floor of 0.0278 (95th percentile 0.0403): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 500 rows (15 equal width bins), against a calibrated-model floor of 0.0278 (95th percentile 0.0403): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 500 rows (15 equal width bins), against a calibrated-model floor of 0.0278 (95th percentile 0.0403): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 500 rows (15 equal width bins), against a calibrated-model floor of 0.0278 (95th percentile 0.0403): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 500 rows (15 equal width bins), against a calibrated-model floor of 0.0278 (95th percentile 0.0403): miscalibration is distinguishable from sampling noise." } }, { "n": 500, "n_bins": 5, "accuracy": 0.3, "mean": 0.029970541953061883, "p95": 0.05014769842329827, "statements": { "0.00125": "ECE 0.0013 over 500 rows (5 equal width bins), against a calibrated-model floor of 0.0300 (95th percentile 0.0501): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 500 rows (5 equal width bins), against a calibrated-model floor of 0.0300 (95th percentile 0.0501): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 500 rows (5 equal width bins), against a calibrated-model floor of 0.0300 (95th percentile 0.0501): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 500 rows (5 equal width bins), against a calibrated-model floor of 0.0300 (95th percentile 0.0501): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 500 rows (5 equal width bins), against a calibrated-model floor of 0.0300 (95th percentile 0.0501): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.074": "ECE 0.0740 over 500 rows (5 equal width bins), against a calibrated-model floor of 0.0300 (95th percentile 0.0501): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 500 rows (5 equal width bins), against a calibrated-model floor of 0.0300 (95th percentile 0.0501): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 500 rows (5 equal width bins), against a calibrated-model floor of 0.0300 (95th percentile 0.0501): miscalibration is distinguishable from sampling noise." } }, { "n": 750, "n_bins": 10, "accuracy": 0.99, "mean": 0.005165671794089274, "p95": 0.00870125465515504, "statements": { "0.00125": "ECE 0.0013 over 750 rows (10 equal width bins), against a calibrated-model floor of 0.0052 (95th percentile 0.0087): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 750 rows (10 equal width bins), against a calibrated-model floor of 0.0052 (95th percentile 0.0087): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 750 rows (10 equal width bins), against a calibrated-model floor of 0.0052 (95th percentile 0.0087): miscalibration is distinguishable from sampling noise.", "0.03125": "ECE 0.0312 over 750 rows (10 equal width bins), against a calibrated-model floor of 0.0052 (95th percentile 0.0087): miscalibration is distinguishable from sampling noise.", "0.05": "ECE 0.0500 over 750 rows (10 equal width bins), against a calibrated-model floor of 0.0052 (95th percentile 0.0087): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 750 rows (10 equal width bins), against a calibrated-model floor of 0.0052 (95th percentile 0.0087): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 750 rows (10 equal width bins), against a calibrated-model floor of 0.0052 (95th percentile 0.0087): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 750 rows (10 equal width bins), against a calibrated-model floor of 0.0052 (95th percentile 0.0087): miscalibration is distinguishable from sampling noise." } }, { "n": 1000, "n_bins": 10, "accuracy": 0.75, "mean": 0.02675736936838183, "p95": 0.04020104887001681, "statements": { "0.00125": "ECE 0.0013 over 1000 rows (10 equal width bins), against a calibrated-model floor of 0.0268 (95th percentile 0.0402): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 1000 rows (10 equal width bins), against a calibrated-model floor of 0.0268 (95th percentile 0.0402): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 1000 rows (10 equal width bins), against a calibrated-model floor of 0.0268 (95th percentile 0.0402): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 1000 rows (10 equal width bins), against a calibrated-model floor of 0.0268 (95th percentile 0.0402): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 1000 rows (10 equal width bins), against a calibrated-model floor of 0.0268 (95th percentile 0.0402): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 1000 rows (10 equal width bins), against a calibrated-model floor of 0.0268 (95th percentile 0.0402): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 1000 rows (10 equal width bins), against a calibrated-model floor of 0.0268 (95th percentile 0.0402): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 1000 rows (10 equal width bins), against a calibrated-model floor of 0.0268 (95th percentile 0.0402): miscalibration is distinguishable from sampling noise." } }, { "n": 1000, "n_bins": 20, "accuracy": 0.85, "mean": 0.029019264436455913, "p95": 0.03978104831245656, "statements": { "0.00125": "ECE 0.0013 over 1000 rows (20 equal width bins), against a calibrated-model floor of 0.0290 (95th percentile 0.0398): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 1000 rows (20 equal width bins), against a calibrated-model floor of 0.0290 (95th percentile 0.0398): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 1000 rows (20 equal width bins), against a calibrated-model floor of 0.0290 (95th percentile 0.0398): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 1000 rows (20 equal width bins), against a calibrated-model floor of 0.0290 (95th percentile 0.0398): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 1000 rows (20 equal width bins), against a calibrated-model floor of 0.0290 (95th percentile 0.0398): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 1000 rows (20 equal width bins), against a calibrated-model floor of 0.0290 (95th percentile 0.0398): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 1000 rows (20 equal width bins), against a calibrated-model floor of 0.0290 (95th percentile 0.0398): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 1000 rows (20 equal width bins), against a calibrated-model floor of 0.0290 (95th percentile 0.0398): miscalibration is distinguishable from sampling noise." } }, { "n": 1000, "n_bins": 50, "accuracy": 0.95, "mean": 0.02521845499263482, "p95": 0.031638820373931995, "statements": { "0.00125": "ECE 0.0013 over 1000 rows (50 equal width bins), against a calibrated-model floor of 0.0252 (95th percentile 0.0316): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 1000 rows (50 equal width bins), against a calibrated-model floor of 0.0252 (95th percentile 0.0316): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 1000 rows (50 equal width bins), against a calibrated-model floor of 0.0252 (95th percentile 0.0316): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 1000 rows (50 equal width bins), against a calibrated-model floor of 0.0252 (95th percentile 0.0316): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.05": "ECE 0.0500 over 1000 rows (50 equal width bins), against a calibrated-model floor of 0.0252 (95th percentile 0.0316): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 1000 rows (50 equal width bins), against a calibrated-model floor of 0.0252 (95th percentile 0.0316): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 1000 rows (50 equal width bins), against a calibrated-model floor of 0.0252 (95th percentile 0.0316): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 1000 rows (50 equal width bins), against a calibrated-model floor of 0.0252 (95th percentile 0.0316): miscalibration is distinguishable from sampling noise." } }, { "n": 2000, "n_bins": 10, "accuracy": 0.8, "mean": 0.017489921485761555, "p95": 0.02596877487096666, "statements": { "0.00125": "ECE 0.0013 over 2000 rows (10 equal width bins), against a calibrated-model floor of 0.0175 (95th percentile 0.0260): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 2000 rows (10 equal width bins), against a calibrated-model floor of 0.0175 (95th percentile 0.0260): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 2000 rows (10 equal width bins), against a calibrated-model floor of 0.0175 (95th percentile 0.0260): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 2000 rows (10 equal width bins), against a calibrated-model floor of 0.0175 (95th percentile 0.0260): miscalibration is distinguishable from sampling noise.", "0.05": "ECE 0.0500 over 2000 rows (10 equal width bins), against a calibrated-model floor of 0.0175 (95th percentile 0.0260): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 2000 rows (10 equal width bins), against a calibrated-model floor of 0.0175 (95th percentile 0.0260): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 2000 rows (10 equal width bins), against a calibrated-model floor of 0.0175 (95th percentile 0.0260): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 2000 rows (10 equal width bins), against a calibrated-model floor of 0.0175 (95th percentile 0.0260): miscalibration is distinguishable from sampling noise." } }, { "n": 2000, "n_bins": 15, "accuracy": 0.1, "mean": 0.013880785016449071, "p95": 0.02054651436977377, "statements": { "0.00125": "ECE 0.0013 over 2000 rows (15 equal width bins), against a calibrated-model floor of 0.0139 (95th percentile 0.0205): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 2000 rows (15 equal width bins), against a calibrated-model floor of 0.0139 (95th percentile 0.0205): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 2000 rows (15 equal width bins), against a calibrated-model floor of 0.0139 (95th percentile 0.0205): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.03125": "ECE 0.0312 over 2000 rows (15 equal width bins), against a calibrated-model floor of 0.0139 (95th percentile 0.0205): miscalibration is distinguishable from sampling noise.", "0.05": "ECE 0.0500 over 2000 rows (15 equal width bins), against a calibrated-model floor of 0.0139 (95th percentile 0.0205): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 2000 rows (15 equal width bins), against a calibrated-model floor of 0.0139 (95th percentile 0.0205): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 2000 rows (15 equal width bins), against a calibrated-model floor of 0.0139 (95th percentile 0.0205): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 2000 rows (15 equal width bins), against a calibrated-model floor of 0.0139 (95th percentile 0.0205): miscalibration is distinguishable from sampling noise." } }, { "n": 4000, "n_bins": 10, "accuracy": 0.8, "mean": 0.012206409389429185, "p95": 0.018615275077065226, "statements": { "0.00125": "ECE 0.0013 over 4000 rows (10 equal width bins), against a calibrated-model floor of 0.0122 (95th percentile 0.0186): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 4000 rows (10 equal width bins), against a calibrated-model floor of 0.0122 (95th percentile 0.0186): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 4000 rows (10 equal width bins), against a calibrated-model floor of 0.0122 (95th percentile 0.0186): miscalibration is distinguishable from sampling noise.", "0.03125": "ECE 0.0312 over 4000 rows (10 equal width bins), against a calibrated-model floor of 0.0122 (95th percentile 0.0186): miscalibration is distinguishable from sampling noise.", "0.05": "ECE 0.0500 over 4000 rows (10 equal width bins), against a calibrated-model floor of 0.0122 (95th percentile 0.0186): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 4000 rows (10 equal width bins), against a calibrated-model floor of 0.0122 (95th percentile 0.0186): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 4000 rows (10 equal width bins), against a calibrated-model floor of 0.0122 (95th percentile 0.0186): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 4000 rows (10 equal width bins), against a calibrated-model floor of 0.0122 (95th percentile 0.0186): miscalibration is distinguishable from sampling noise." } }, { "n": 5000, "n_bins": 20, "accuracy": 0.97, "mean": 0.005268464475778579, "p95": 0.00761369266481136, "statements": { "0.00125": "ECE 0.0013 over 5000 rows (20 equal width bins), against a calibrated-model floor of 0.0053 (95th percentile 0.0076): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 5000 rows (20 equal width bins), against a calibrated-model floor of 0.0053 (95th percentile 0.0076): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 5000 rows (20 equal width bins), against a calibrated-model floor of 0.0053 (95th percentile 0.0076): miscalibration is distinguishable from sampling noise.", "0.03125": "ECE 0.0312 over 5000 rows (20 equal width bins), against a calibrated-model floor of 0.0053 (95th percentile 0.0076): miscalibration is distinguishable from sampling noise.", "0.05": "ECE 0.0500 over 5000 rows (20 equal width bins), against a calibrated-model floor of 0.0053 (95th percentile 0.0076): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 5000 rows (20 equal width bins), against a calibrated-model floor of 0.0053 (95th percentile 0.0076): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 5000 rows (20 equal width bins), against a calibrated-model floor of 0.0053 (95th percentile 0.0076): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 5000 rows (20 equal width bins), against a calibrated-model floor of 0.0053 (95th percentile 0.0076): miscalibration is distinguishable from sampling noise." } }, { "n": 10000, "n_bins": 10, "accuracy": 0.8, "mean": 0.007729592376046467, "p95": 0.011776573914467762, "statements": { "0.00125": "ECE 0.0013 over 10000 rows (10 equal width bins), against a calibrated-model floor of 0.0077 (95th percentile 0.0118): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 10000 rows (10 equal width bins), against a calibrated-model floor of 0.0077 (95th percentile 0.0118): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 10000 rows (10 equal width bins), against a calibrated-model floor of 0.0077 (95th percentile 0.0118): miscalibration is distinguishable from sampling noise.", "0.03125": "ECE 0.0312 over 10000 rows (10 equal width bins), against a calibrated-model floor of 0.0077 (95th percentile 0.0118): miscalibration is distinguishable from sampling noise.", "0.05": "ECE 0.0500 over 10000 rows (10 equal width bins), against a calibrated-model floor of 0.0077 (95th percentile 0.0118): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 10000 rows (10 equal width bins), against a calibrated-model floor of 0.0077 (95th percentile 0.0118): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 10000 rows (10 equal width bins), against a calibrated-model floor of 0.0077 (95th percentile 0.0118): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 10000 rows (10 equal width bins), against a calibrated-model floor of 0.0077 (95th percentile 0.0118): miscalibration is distinguishable from sampling noise." } }, { "n": 10000, "n_bins": 30, "accuracy": 0.9, "mean": 0.008988447318912384, "p95": 0.011866513346387128, "statements": { "0.00125": "ECE 0.0013 over 10000 rows (30 equal width bins), against a calibrated-model floor of 0.0090 (95th percentile 0.0119): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 10000 rows (30 equal width bins), against a calibrated-model floor of 0.0090 (95th percentile 0.0119): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 10000 rows (30 equal width bins), against a calibrated-model floor of 0.0090 (95th percentile 0.0119): miscalibration is distinguishable from sampling noise.", "0.03125": "ECE 0.0312 over 10000 rows (30 equal width bins), against a calibrated-model floor of 0.0090 (95th percentile 0.0119): miscalibration is distinguishable from sampling noise.", "0.05": "ECE 0.0500 over 10000 rows (30 equal width bins), against a calibrated-model floor of 0.0090 (95th percentile 0.0119): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 10000 rows (30 equal width bins), against a calibrated-model floor of 0.0090 (95th percentile 0.0119): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 10000 rows (30 equal width bins), against a calibrated-model floor of 0.0090 (95th percentile 0.0119): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 10000 rows (30 equal width bins), against a calibrated-model floor of 0.0090 (95th percentile 0.0119): miscalibration is distinguishable from sampling noise." } }, { "n": 20000, "n_bins": 10, "accuracy": 0.65, "mean": 0.006857970527670982, "p95": 0.010249792847569345, "statements": { "0.00125": "ECE 0.0013 over 20000 rows (10 equal width bins), against a calibrated-model floor of 0.0069 (95th percentile 0.0102): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.005": "ECE 0.0050 over 20000 rows (10 equal width bins), against a calibrated-model floor of 0.0069 (95th percentile 0.0102): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable.", "0.02": "ECE 0.0200 over 20000 rows (10 equal width bins), against a calibrated-model floor of 0.0069 (95th percentile 0.0102): miscalibration is distinguishable from sampling noise.", "0.03125": "ECE 0.0312 over 20000 rows (10 equal width bins), against a calibrated-model floor of 0.0069 (95th percentile 0.0102): miscalibration is distinguishable from sampling noise.", "0.05": "ECE 0.0500 over 20000 rows (10 equal width bins), against a calibrated-model floor of 0.0069 (95th percentile 0.0102): miscalibration is distinguishable from sampling noise.", "0.074": "ECE 0.0740 over 20000 rows (10 equal width bins), against a calibrated-model floor of 0.0069 (95th percentile 0.0102): miscalibration is distinguishable from sampling noise.", "0.1": "ECE 0.1000 over 20000 rows (10 equal width bins), against a calibrated-model floor of 0.0069 (95th percentile 0.0102): miscalibration is distinguishable from sampling noise.", "0.2": "ECE 0.2000 over 20000 rows (10 equal width bins), against a calibrated-model floor of 0.0069 (95th percentile 0.0102): miscalibration is distinguishable from sampling noise." } } ], "fixed4": { "4.9999999999999996e-05": "0.0000", "5e-05": "0.0001", "5.000000000000001e-05": "0.0001", "0.0007499999999999999": "0.0007", "0.00075": "0.0008", "0.0007500000000000001": "0.0008", "0.0014499999999999997": "0.0014", "0.00145": "0.0014", "0.0014500000000000001": "0.0015", "0.0021499999999999996": "0.0021", "0.00215": "0.0022", "0.0021500000000000004": "0.0022", "0.0028499999999999997": "0.0028", "0.00285": "0.0029", "0.0028500000000000005": "0.0029", "0.0035499999999999998": "0.0035", "0.00355": "0.0036", "0.0035500000000000006": "0.0036", "0.0042499999999999994": "0.0042", "0.00425": "0.0043", "0.004250000000000001": "0.0043", "0.0049499999999999995": "0.0049", "0.00495": "0.0050", "0.004950000000000001": "0.0050", "0.005649999999999999": "0.0056", "0.00565": "0.0056", "0.0056500000000000005": "0.0057", "0.006349999999999999": "0.0063", "0.00635": "0.0063", "0.006350000000000001": "0.0064", "0.007049999999999999": "0.0070", "0.00705": "0.0070", "0.007050000000000001": "0.0071", "0.007749999999999999": "0.0077", "0.00775": "0.0077", "0.007750000000000001": "0.0078", "0.008449999999999997": "0.0084", "0.00845": "0.0084", "0.008450000000000001": "0.0085", "0.009149999999999998": "0.0091", "0.00915": "0.0092", "0.009150000000000002": "0.0092", "0.009849999999999998": "0.0098", "0.00985": "0.0098", "0.009850000000000001": "0.0099", "0.010549999999999999": "0.0105", "0.01055": "0.0106", "0.010550000000000002": "0.0106", "0.011249999999999998": "0.0112", "0.01125": "0.0112", "0.011250000000000001": "0.0113", "0.011949999999999999": "0.0119", "0.01195": "0.0120", "0.011950000000000002": "0.0120", "0.012649999999999998": "0.0126", "0.01265": "0.0126", "0.012650000000000002": "0.0127", "0.013349999999999999": "0.0133", "0.01335": "0.0134", "0.013350000000000002": "0.0134", "0.014049999999999998": "0.0140", "0.01405": "0.0140", "0.014050000000000002": "0.0141", "0.014749999999999997": "0.0147", "0.01475": "0.0147", "0.014750000000000001": "0.0148", "0.015449999999999998": "0.0154", "0.01545": "0.0155", "0.015450000000000002": "0.0155", "0.016149999999999998": "0.0161", "0.01615": "0.0162", "0.016150000000000005": "0.0162", "0.016849999999999997": "0.0168", "0.01685": "0.0169", "0.016850000000000004": "0.0169", "0.017549999999999996": "0.0175", "0.01755": "0.0175", "0.017550000000000003": "0.0176", "0.018249999999999995": "0.0182", "0.01825": "0.0182", "0.018250000000000002": "0.0183", "0.018949999999999998": "0.0189", "0.01895": "0.0190", "0.018950000000000005": "0.0190", "0.019649999999999997": "0.0196", "0.01965": "0.0197", "0.019650000000000004": "0.0197", "0.020349999999999997": "0.0203", "0.02035": "0.0204", "0.020350000000000004": "0.0204", "0.021049999999999996": "0.0210", "0.02105": "0.0210", "0.021050000000000003": "0.0211", "0.021749999999999995": "0.0217", "0.02175": "0.0217", "0.021750000000000002": "0.0218", "0.022449999999999998": "0.0224", "0.02245": "0.0225", "0.022450000000000005": "0.0225", "0.023149999999999997": "0.0231", "0.02315": "0.0232", "0.023150000000000004": "0.0232", "0.023849999999999996": "0.0238", "0.02385": "0.0238", "0.023850000000000003": "0.0239", "0.024549999999999995": "0.0245", "0.02455": "0.0245", "0.024550000000000002": "0.0246", "0.025249999999999998": "0.0252", "0.02525": "0.0253", "0.025250000000000005": "0.0253", "0.025949999999999997": "0.0259", "0.02595": "0.0260", "0.025950000000000004": "0.0260", "0.026649999999999997": "0.0266", "0.02665": "0.0267", "0.026650000000000004": "0.0267", "0.027349999999999996": "0.0273", "0.02735": "0.0273", "0.027350000000000003": "0.0274", "0.028049999999999995": "0.0280", "0.02805": "0.0280", "0.028050000000000002": "0.0281", "0.028749999999999998": "0.0287", "0.02875": "0.0288", "0.028750000000000005": "0.0288", "0.029449999999999997": "0.0294", "0.02945": "0.0295", "0.029450000000000004": "0.0295", "0.030149999999999996": "0.0301", "0.03015": "0.0301", "0.030150000000000003": "0.0302", "0.030849999999999995": "0.0308", "0.03085": "0.0308", "0.030850000000000002": "0.0309", "0.03125": "0.0312", "0.031549999999999995": "0.0315", "0.03155": "0.0316", "0.03155000000000001": "0.0316", "0.032249999999999994": "0.0322", "0.03225": "0.0323", "0.03225000000000001": "0.0323", "0.03294999999999999": "0.0329", "0.03295": "0.0330", "0.03295000000000001": "0.0330", "0.03364999999999999": "0.0336", "0.03365": "0.0336", "0.033650000000000006": "0.0337", "0.03434999999999999": "0.0343", "0.03435": "0.0343", "0.034350000000000006": "0.0344", "0.03504999999999999": "0.0350", "0.03505": "0.0350", "0.035050000000000005": "0.0351", "0.03574999999999999": "0.0357", "0.03575": "0.0357", "0.035750000000000004": "0.0358", "0.036449999999999996": "0.0364", "0.03645": "0.0365", "0.03645000000000001": "0.0365", "0.037149999999999996": "0.0371", "0.03715": "0.0372", "0.03715000000000001": "0.0372", "0.037849999999999995": "0.0378", "0.03785": "0.0379", "0.03785000000000001": "0.0379", "0.038549999999999994": "0.0385", "0.03855": "0.0386", "0.03855000000000001": "0.0386", "0.03924999999999999": "0.0392", "0.03925": "0.0393", "0.03925000000000001": "0.0393", "0.03994999999999999": "0.0399", "0.03995": "0.0399", "0.039950000000000006": "0.0400", "0.04064999999999999": "0.0406", "0.04065": "0.0406", "0.040650000000000006": "0.0407", "0.04134999999999999": "0.0413", "0.04135": "0.0413", "0.041350000000000005": "0.0414", "0.04204999999999999": "0.0420", "0.04205": "0.0420", "0.042050000000000004": "0.0421", "0.042749999999999996": "0.0427", "0.04275": "0.0428", "0.04275000000000001": "0.0428", "0.043449999999999996": "0.0434", "0.04345": "0.0435", "0.04345000000000001": "0.0435", "0.044149999999999995": "0.0441", "0.04415": "0.0442", "0.04415000000000001": "0.0442", "0.044849999999999994": "0.0448", "0.04485": "0.0449", "0.04485000000000001": "0.0449", "0.04554999999999999": "0.0455", "0.04555": "0.0456", "0.04555000000000001": "0.0456", "0.04624999999999999": "0.0462", "0.04625": "0.0462", "0.046250000000000006": "0.0463", "0.04694999999999999": "0.0469", "0.04695": "0.0469", "0.046950000000000006": "0.0470", "0.04764999999999999": "0.0476", "0.04765": "0.0476", "0.047650000000000005": "0.0477", "0.04834999999999999": "0.0483", "0.04835": "0.0483", "0.048350000000000004": "0.0484", "0.049049999999999996": "0.0490", "0.04905": "0.0491", "0.04905000000000001": "0.0491", "0.049749999999999996": "0.0497", "0.04975": "0.0498", "0.04975000000000001": "0.0498", "0.050449999999999995": "0.0504", "0.05045": "0.0505", "0.05045000000000001": "0.0505", "0.051149999999999994": "0.0511", "0.05115": "0.0512", "0.05115000000000001": "0.0512", "0.05184999999999999": "0.0518", "0.05185": "0.0519", "0.05185000000000001": "0.0519", "0.05254999999999999": "0.0525", "0.05255": "0.0525", "0.052550000000000006": "0.0526", "0.05324999999999999": "0.0532", "0.05325": "0.0532", "0.053250000000000006": "0.0533", "0.05394999999999999": "0.0539", "0.05395": "0.0539", "0.053950000000000005": "0.0540", "0.05464999999999999": "0.0546", "0.05465": "0.0546", "0.054650000000000004": "0.0547", "0.055349999999999996": "0.0553", "0.05535": "0.0554", "0.05535000000000001": "0.0554", "0.056049999999999996": "0.0560", "0.05605": "0.0561", "0.05605000000000001": "0.0561", "0.056749999999999995": "0.0567", "0.05675": "0.0568", "0.05675000000000001": "0.0568", "0.057449999999999994": "0.0574", "0.05745": "0.0575", "0.05745000000000001": "0.0575", "0.05814999999999999": "0.0581", "0.05815": "0.0582", "0.05815000000000001": "0.0582", "0.05884999999999999": "0.0588", "0.05885": "0.0588", "0.058850000000000006": "0.0589", "0.05954999999999999": "0.0595", "0.05955": "0.0595", "0.059550000000000006": "0.0596", "0.06024999999999999": "0.0602", "0.06025": "0.0602", "0.060250000000000005": "0.0603", "0.06094999999999999": "0.0609", "0.06095": "0.0609", "0.060950000000000004": "0.0610", "0.061649999999999996": "0.0616", "0.06165": "0.0617", "0.06165000000000001": "0.0617", "0.062349999999999996": "0.0623", "0.06235": "0.0624", "0.06235000000000001": "0.0624", "0.06304999999999998": "0.0630", "0.06305": "0.0630", "0.06305000000000001": "0.0631", "0.06374999999999999": "0.0637", "0.06375": "0.0638", "0.06375000000000001": "0.0638", "0.06444999999999998": "0.0644", "0.06445": "0.0644", "0.06445000000000001": "0.0645", "0.06514999999999999": "0.0651", "0.06515": "0.0651", "0.06515000000000001": "0.0652", "0.06584999999999999": "0.0658", "0.06585": "0.0659", "0.06585000000000002": "0.0659", "0.06654999999999998": "0.0665", "0.06655": "0.0665", "0.06655000000000001": "0.0666", "0.06724999999999999": "0.0672", "0.06725": "0.0673", "0.06725000000000002": "0.0673", "0.06794999999999998": "0.0679", "0.06795": "0.0679", "0.06795000000000001": "0.0680", "0.06864999999999999": "0.0686", "0.06865": "0.0687", "0.06865000000000002": "0.0687", "0.06934999999999998": "0.0693", "0.06935": "0.0693", "0.06935000000000001": "0.0694", "0.07004999999999999": "0.0700", "0.07005": "0.0701", "0.07005000000000002": "0.0701", "0.07074999999999998": "0.0707", "0.07075": "0.0707", "0.07075000000000001": "0.0708", "0.07144999999999999": "0.0714", "0.07145": "0.0714", "0.07145000000000001": "0.0715", "0.07214999999999999": "0.0721", "0.07215": "0.0722", "0.07215000000000002": "0.0722", "0.07284999999999998": "0.0728", "0.07285": "0.0728", "0.07285000000000001": "0.0729", "0.07354999999999999": "0.0735", "0.07355": "0.0736", "0.07355000000000002": "0.0736", "0.07424999999999998": "0.0742", "0.07425": "0.0742", "0.07425000000000001": "0.0743", "0.07494999999999999": "0.0749", "0.07495": "0.0750", "0.07495000000000002": "0.0750", "0.07564999999999998": "0.0756", "0.07565": "0.0756", "0.07565000000000001": "0.0757", "0.07634999999999999": "0.0763", "0.07635": "0.0764", "0.07635000000000002": "0.0764", "0.07704999999999998": "0.0770", "0.07705": "0.0770", "0.07705000000000001": "0.0771", "0.07774999999999999": "0.0777", "0.07775": "0.0777", "0.07775000000000001": "0.0778", "0.07844999999999999": "0.0784", "0.07845": "0.0785", "0.07845000000000002": "0.0785", "0.07914999999999998": "0.0791", "0.07915": "0.0791", "0.07915000000000001": "0.0792", "0.07984999999999999": "0.0798", "0.07985": "0.0799", "0.07985000000000002": "0.0799", "0.08054999999999998": "0.0805", "0.08055": "0.0805", "0.08055000000000001": "0.0806", "0.08124999999999999": "0.0812", "0.08125": "0.0813", "0.08125000000000002": "0.0813", "0.08194999999999998": "0.0819", "0.08195": "0.0819", "0.08195000000000001": "0.0820", "0.08264999999999999": "0.0826", "0.08265": "0.0827", "0.08265000000000002": "0.0827", "0.08334999999999998": "0.0833", "0.08335": "0.0833", "0.08335000000000001": "0.0834", "0.08404999999999999": "0.0840", "0.08405": "0.0840", "0.08405000000000001": "0.0841", "0.08474999999999999": "0.0847", "0.08475": "0.0848", "0.08475000000000002": "0.0848", "0.08544999999999998": "0.0854", "0.08545": "0.0854", "0.08545000000000001": "0.0855", "0.08614999999999999": "0.0861", "0.08615": "0.0862", "0.08615000000000002": "0.0862", "0.08684999999999998": "0.0868", "0.08685": "0.0868", "0.08685000000000001": "0.0869", "0.08754999999999999": "0.0875", "0.08755": "0.0876", "0.08755000000000002": "0.0876", "0.08824999999999998": "0.0882", "0.08825": "0.0882", "0.08825000000000001": "0.0883", "0.08894999999999999": "0.0889", "0.08895": "0.0890", "0.08895000000000002": "0.0890", "0.08964999999999998": "0.0896", "0.08965": "0.0896", "0.08965000000000001": "0.0897", "0.09034999999999999": "0.0903", "0.09035": "0.0903", "0.09035000000000001": "0.0904", "0.09104999999999999": "0.0910", "0.09105": "0.0911", "0.09105000000000002": "0.0911", "0.09174999999999998": "0.0917", "0.09175": "0.0917", "0.09175000000000001": "0.0918", "0.09244999999999999": "0.0924", "0.09245": "0.0925", "0.09245000000000002": "0.0925", "0.09314999999999998": "0.0931", "0.09315": "0.0931", "0.09315000000000001": "0.0932", "0.09375": "0.0938", "0.09384999999999999": "0.0938", "0.09385": "0.0939", "0.09385000000000002": "0.0939", "0.09454999999999998": "0.0945", "0.09455": "0.0945", "0.09455000000000001": "0.0946", "0.09524999999999999": "0.0952", "0.09525": "0.0953", "0.09525000000000002": "0.0953", "0.09594999999999998": "0.0959", "0.09595": "0.0959", "0.09595000000000001": "0.0960", "0.09664999999999999": "0.0966", "0.09665": "0.0966", "0.09665000000000001": "0.0967", "0.09734999999999999": "0.0973", "0.09735": "0.0974", "0.09735000000000002": "0.0974", "0.09804999999999998": "0.0980", "0.09805": "0.0980", "0.09805000000000001": "0.0981", "0.09874999999999999": "0.0987", "0.09875": "0.0988", "0.09875000000000002": "0.0988", "0.09944999999999998": "0.0994", "0.09945": "0.0994", "0.09945000000000001": "0.0995", "0.10014999999999999": "0.1001", "0.10015": "0.1002", "0.10015000000000002": "0.1002", "0.10084999999999998": "0.1008", "0.10085": "0.1008", "0.10085000000000001": "0.1009", "0.10154999999999999": "0.1015", "0.10155": "0.1016", "0.10155000000000002": "0.1016", "0.10224999999999998": "0.1022", "0.10225": "0.1022", "0.10225000000000001": "0.1023", "0.10294999999999999": "0.1029", "0.10295": "0.1029", "0.10295000000000001": "0.1030", "0.10364999999999999": "0.1036", "0.10365": "0.1037", "0.10365000000000002": "0.1037", "0.10434999999999998": "0.1043", "0.10435": "0.1043", "0.10435000000000001": "0.1044", "0.10504999999999999": "0.1050", "0.10505": "0.1051", "0.10505000000000002": "0.1051", "0.10574999999999998": "0.1057", "0.10575": "0.1057", "0.10575000000000001": "0.1058", "0.10644999999999999": "0.1064", "0.10645": "0.1065", "0.10645000000000002": "0.1065", "0.10714999999999998": "0.1071", "0.10715": "0.1071", "0.10715000000000001": "0.1072", "0.10784999999999999": "0.1078", "0.10785": "0.1079", "0.10785000000000002": "0.1079", "0.10854999999999998": "0.1085", "0.10855": "0.1085", "0.10855000000000001": "0.1086", "0.10924999999999999": "0.1092", "0.10925": "0.1092", "0.10925000000000001": "0.1093", "0.10994999999999999": "0.1099", "0.10995": "0.1100", "0.10995000000000002": "0.1100", "0.11064999999999998": "0.1106", "0.11065": "0.1106", "0.11065000000000001": "0.1107", "0.11134999999999999": "0.1113", "0.11135": "0.1114", "0.11135000000000002": "0.1114", "0.11204999999999998": "0.1120", "0.11205": "0.1120", "0.11205000000000001": "0.1121", "0.11274999999999999": "0.1127", "0.11275": "0.1128", "0.11275000000000002": "0.1128", "0.11344999999999998": "0.1134", "0.11345": "0.1134", "0.11345000000000001": "0.1135", "0.11414999999999999": "0.1141", "0.11415": "0.1142", "0.11415000000000002": "0.1142", "0.11484999999999998": "0.1148", "0.11485": "0.1148", "0.11485000000000001": "0.1149", "0.11554999999999999": "0.1155", "0.11555": "0.1155", "0.11555000000000001": "0.1156", "0.11624999999999999": "0.1162", "0.11625": "0.1163", "0.11625000000000002": "0.1163", "0.11694999999999998": "0.1169", "0.11695": "0.1169", "0.11695000000000001": "0.1170", "0.11764999999999999": "0.1176", "0.11765": "0.1177", "0.11765000000000002": "0.1177", "0.11834999999999998": "0.1183", "0.11835": "0.1183", "0.11835000000000001": "0.1184", "0.11904999999999999": "0.1190", "0.11905": "0.1191", "0.11905000000000002": "0.1191", "0.11974999999999998": "0.1197", "0.11975": "0.1197", "0.11975000000000001": "0.1198", "0.12044999999999999": "0.1204", "0.12045": "0.1205", "0.12045000000000002": "0.1205", "0.12114999999999998": "0.1211", "0.12115": "0.1211", "0.12115000000000001": "0.1212", "0.12184999999999999": "0.1218", "0.12185": "0.1218", "0.12185000000000001": "0.1219", "0.12254999999999999": "0.1225", "0.12255": "0.1226", "0.12255000000000002": "0.1226", "0.12324999999999998": "0.1232", "0.12325": "0.1232", "0.12325000000000001": "0.1233", "0.12394999999999999": "0.1239", "0.12395": "0.1240", "0.12395000000000002": "0.1240", "0.12464999999999998": "0.1246", "0.12465": "0.1246", "0.12465000000000001": "0.1247", "0.12534999999999996": "0.1253", "0.12535": "0.1253", "0.12535000000000002": "0.1254", "0.12604999999999997": "0.1260", "0.12605": "0.1260", "0.12605000000000002": "0.1261", "0.12674999999999997": "0.1267", "0.12675": "0.1268", "0.12675000000000003": "0.1268", "0.12744999999999998": "0.1274", "0.12745": "0.1275", "0.12745000000000004": "0.1275", "0.12814999999999996": "0.1281", "0.12815": "0.1281", "0.12815000000000001": "0.1282", "0.12884999999999996": "0.1288", "0.12885": "0.1288", "0.12885000000000002": "0.1289", "0.12954999999999997": "0.1295", "0.12955": "0.1295", "0.12955000000000003": "0.1296", "0.13024999999999998": "0.1302", "0.13025": "0.1303", "0.13025000000000003": "0.1303", "0.13094999999999998": "0.1309", "0.13095": "0.1310", "0.13095000000000004": "0.1310", "0.13164999999999996": "0.1316", "0.13165": "0.1316", "0.13165000000000002": "0.1317", "0.13234999999999997": "0.1323", "0.13235": "0.1323", "0.13235000000000002": "0.1324", "0.13304999999999997": "0.1330", "0.13305": "0.1331", "0.13305000000000003": "0.1331", "0.13374999999999998": "0.1337", "0.13375": "0.1338", "0.13375000000000004": "0.1338", "0.13444999999999996": "0.1344", "0.13445": "0.1344", "0.13445000000000001": "0.1345", "0.13514999999999996": "0.1351", "0.13515": "0.1351", "0.13515000000000002": "0.1352", "0.13584999999999997": "0.1358", "0.13585": "0.1358", "0.13585000000000003": "0.1359", "0.13654999999999998": "0.1365", "0.13655": "0.1366", "0.13655000000000003": "0.1366", "0.13724999999999998": "0.1372", "0.13725": "0.1373", "0.13725000000000004": "0.1373", "0.13794999999999996": "0.1379", "0.13795": "0.1379", "0.13795000000000002": "0.1380", "0.13864999999999997": "0.1386", "0.13865": "0.1386", "0.13865000000000002": "0.1387", "0.13934999999999997": "0.1393", "0.13935": "0.1394", "0.13935000000000003": "0.1394", "0.14004999999999998": "0.1400", "0.14005": "0.1401", "0.14005000000000004": "0.1401", "0.14074999999999996": "0.1407", "0.14075": "0.1407", "0.14075000000000001": "0.1408", "0.14144999999999996": "0.1414", "0.14145": "0.1414", "0.14145000000000002": "0.1415", "0.14214999999999997": "0.1421", "0.14215": "0.1421", "0.14215000000000003": "0.1422", "0.14284999999999998": "0.1428", "0.14285": "0.1429", "0.14285000000000003": "0.1429", "0.14354999999999998": "0.1435", "0.14355": "0.1436", "0.14355000000000004": "0.1436", "0.14424999999999996": "0.1442", "0.14425": "0.1442", "0.14425000000000002": "0.1443", "0.14494999999999997": "0.1449", "0.14495": "0.1449", "0.14495000000000002": "0.1450", "0.14564999999999997": "0.1456", "0.14565": "0.1457", "0.14565000000000003": "0.1457", "0.14634999999999998": "0.1463", "0.14635": "0.1464", "0.14635000000000004": "0.1464", "0.14704999999999996": "0.1470", "0.14705": "0.1470", "0.14705000000000001": "0.1471", "0.14774999999999996": "0.1477", "0.14775": "0.1477", "0.14775000000000002": "0.1478", "0.14844999999999997": "0.1484", "0.14845": "0.1484", "0.14845000000000003": "0.1485", "0.14914999999999998": "0.1491", "0.14915": "0.1492", "0.14915000000000003": "0.1492", "0.14984999999999998": "0.1498", "0.14985": "0.1499", "0.14985000000000004": "0.1499", "0.15054999999999996": "0.1505", "0.15055": "0.1505", "0.15055000000000002": "0.1506", "0.15124999999999997": "0.1512", "0.15125": "0.1512", "0.15125000000000002": "0.1513", "0.15194999999999997": "0.1519", "0.15195": "0.1520", "0.15195000000000003": "0.1520", "0.15264999999999998": "0.1526", "0.15265": "0.1527", "0.15265000000000004": "0.1527", "0.15334999999999996": "0.1533", "0.15335": "0.1533", "0.15335000000000001": "0.1534", "0.15404999999999996": "0.1540", "0.15405": "0.1540", "0.15405000000000002": "0.1541", "0.15474999999999997": "0.1547", "0.15475": "0.1547", "0.15475000000000003": "0.1548", "0.15544999999999998": "0.1554", "0.15545": "0.1555", "0.15545000000000003": "0.1555", "0.15614999999999998": "0.1561", "0.15615": "0.1562", "0.15615000000000004": "0.1562", "0.15625": "0.1562", "0.15684999999999996": "0.1568", "0.15685": "0.1568", "0.15685000000000002": "0.1569", "0.15754999999999997": "0.1575", "0.15755": "0.1575", "0.15755000000000002": "0.1576", "0.15824999999999997": "0.1582", "0.15825": "0.1583", "0.15825000000000003": "0.1583", "0.15894999999999998": "0.1589", "0.15895": "0.1590", "0.15895000000000004": "0.1590", "0.15964999999999996": "0.1596", "0.15965": "0.1596", "0.15965000000000001": "0.1597", "0.16034999999999996": "0.1603", "0.16035": "0.1603", "0.16035000000000002": "0.1604", "0.16104999999999997": "0.1610", "0.16105": "0.1610", "0.16105000000000003": "0.1611", "0.16174999999999998": "0.1617", "0.16175": "0.1618", "0.16175000000000003": "0.1618", "0.16244999999999998": "0.1624", "0.16245": "0.1625", "0.16245000000000004": "0.1625", "0.16314999999999996": "0.1631", "0.16315": "0.1631", "0.16315000000000002": "0.1632", "0.16384999999999997": "0.1638", "0.16385": "0.1638", "0.16385000000000002": "0.1639", "0.16454999999999997": "0.1645", "0.16455": "0.1646", "0.16455000000000003": "0.1646", "0.16524999999999998": "0.1652", "0.16525": "0.1653", "0.16525000000000004": "0.1653", "0.16594999999999996": "0.1659", "0.16595": "0.1659", "0.16595000000000001": "0.1660", "0.16664999999999996": "0.1666", "0.16665": "0.1666", "0.16665000000000002": "0.1667", "0.16734999999999997": "0.1673", "0.16735": "0.1673", "0.16735000000000003": "0.1674", "0.16804999999999998": "0.1680", "0.16805": "0.1681", "0.16805000000000003": "0.1681", "0.16874999999999998": "0.1687", "0.16875": "0.1688", "0.16875000000000004": "0.1688", "0.16944999999999996": "0.1694", "0.16945": "0.1694", "0.16945000000000002": "0.1695", "0.17014999999999997": "0.1701", "0.17015": "0.1701", "0.17015000000000002": "0.1702", "0.17084999999999997": "0.1708", "0.17085": "0.1709", "0.17085000000000003": "0.1709", "0.17154999999999998": "0.1715", "0.17155": "0.1716", "0.17155000000000004": "0.1716", "0.17224999999999996": "0.1722", "0.17225": "0.1722", "0.17225000000000001": "0.1723", "0.17294999999999996": "0.1729", "0.17295": "0.1729", "0.17295000000000002": "0.1730", "0.17364999999999997": "0.1736", "0.17365": "0.1736", "0.17365000000000003": "0.1737", "0.17434999999999998": "0.1743", "0.17435": "0.1744", "0.17435000000000003": "0.1744", "0.17504999999999998": "0.1750", "0.17505": "0.1751", "0.17505000000000004": "0.1751", "0.17574999999999996": "0.1757", "0.17575": "0.1757", "0.17575000000000002": "0.1758", "0.17644999999999997": "0.1764", "0.17645": "0.1764", "0.17645000000000002": "0.1765", "0.17714999999999997": "0.1771", "0.17715": "0.1772", "0.17715000000000003": "0.1772", "0.17784999999999998": "0.1778", "0.17785": "0.1779", "0.17785000000000004": "0.1779", "0.17854999999999996": "0.1785", "0.17855": "0.1785", "0.17855000000000001": "0.1786", "0.17924999999999996": "0.1792", "0.17925": "0.1792", "0.17925000000000002": "0.1793", "0.17994999999999997": "0.1799", "0.17995": "0.1799", "0.17995000000000003": "0.1800", "0.18064999999999998": "0.1806", "0.18065": "0.1807", "0.18065000000000003": "0.1807", "0.18134999999999998": "0.1813", "0.18135": "0.1814", "0.18135000000000004": "0.1814", "0.18204999999999996": "0.1820", "0.18205": "0.1820", "0.18205000000000002": "0.1821", "0.18274999999999997": "0.1827", "0.18275": "0.1827", "0.18275000000000002": "0.1828", "0.18344999999999997": "0.1834", "0.18345": "0.1835", "0.18345000000000003": "0.1835", "0.18414999999999998": "0.1841", "0.18415": "0.1842", "0.18415000000000004": "0.1842", "0.18484999999999996": "0.1848", "0.18485": "0.1848", "0.18485000000000001": "0.1849", "0.18554999999999996": "0.1855", "0.18555": "0.1855", "0.18555000000000002": "0.1856", "0.18624999999999997": "0.1862", "0.18625": "0.1862", "0.18625000000000003": "0.1863", "0.18694999999999998": "0.1869", "0.18695": "0.1870", "0.18695000000000003": "0.1870", "0.18764999999999998": "0.1876", "0.18765": "0.1877", "0.18765000000000004": "0.1877", "0.18834999999999996": "0.1883", "0.18835": "0.1883", "0.18835000000000002": "0.1884", "0.18904999999999997": "0.1890", "0.18905": "0.1890", "0.18905000000000002": "0.1891", "0.18974999999999997": "0.1897", "0.18975": "0.1898", "0.18975000000000003": "0.1898", "0.19044999999999998": "0.1904", "0.19045": "0.1905", "0.19045000000000004": "0.1905", "0.19114999999999996": "0.1911", "0.19115": "0.1911", "0.19115000000000001": "0.1912", "0.19184999999999997": "0.1918", "0.19185": "0.1918", "0.19185000000000002": "0.1919", "0.19254999999999997": "0.1925", "0.19255": "0.1925", "0.19255000000000003": "0.1926", "0.19324999999999998": "0.1932", "0.19325": "0.1933", "0.19325000000000003": "0.1933", "0.19394999999999998": "0.1939", "0.19395": "0.1940", "0.19395000000000004": "0.1940", "0.19464999999999996": "0.1946", "0.19465": "0.1946", "0.19465000000000002": "0.1947", "0.19534999999999997": "0.1953", "0.19535": "0.1953", "0.19535000000000002": "0.1954", "0.19604999999999997": "0.1960", "0.19605": "0.1961", "0.19605000000000003": "0.1961", "0.19674999999999998": "0.1967", "0.19675": "0.1968", "0.19675000000000004": "0.1968", "0.19744999999999996": "0.1974", "0.19745": "0.1974", "0.19745000000000001": "0.1975", "0.19814999999999997": "0.1981", "0.19815": "0.1981", "0.19815000000000002": "0.1982", "0.19884999999999997": "0.1988", "0.19885": "0.1988", "0.19885000000000003": "0.1989", "0.19954999999999998": "0.1995", "0.19955": "0.1996", "0.19955000000000003": "0.1996", "0.21875": "0.2188", "0.28125": "0.2812", "0.34375": "0.3438", "0.40625": "0.4062", "0.46875": "0.4688", "0.53125": "0.5312", "0.59375": "0.5938", "0.65625": "0.6562", "0.71875": "0.7188", "0.78125": "0.7812", "0.84375": "0.8438", "0.90625": "0.9062", "0.96875": "0.9688", "1.03125": "1.0312", "1.09375": "1.0938", "1.15625": "1.1562", "1.21875": "1.2188", "1.28125": "1.2812", "1.34375": "1.3438", "1.40625": "1.4062", "1.46875": "1.4688", "1.53125": "1.5312", "1.59375": "1.5938", "1.65625": "1.6562", "1.71875": "1.7188", "1.78125": "1.7812", "1.84375": "1.8438", "1.90625": "1.9062", "1.96875": "1.9688" }, "pasted": [ { "label": "comma separated, with a header", "n_bins": 10, "text": "probability,outcome\n0.3701902842671547,0\n0.6636748580851755,0\n0.4854677254876745,1\n0.9809095860229241,1\n0.5402041081123293,1\n0.5878420282243031,1\n0.6429783201852302,1\n0.6914959785439542,1\n0.8429774994285216,1\n0.6053664379951705,0\n0.5505226687317148,1\n0.47873665214418765,1\n0.9186624277149896,0\n0.32830736621921897,1\n0.8046851732099684,0\n0.8983728404473525,1\n0.7176103346111447,0\n0.6845520598937217,1\n0.8754085153219575,1\n0.8274990808507586,1\n0.8165098359026878,0\n0.7370462486190534,1\n0.8136063683786394,1\n0.8120125108755948,1\n0.7847554230365376,1\n0.7890675244463915,1\n0.5442676637935917,1\n0.7571337761499041,1\n0.9021548876500897,1\n0.637880062408436,1\n0.7557549754401213,1\n0.5083763821624885,1\n0.9141689143748283,1\n0.8253786551448449,1\n0.9130937777894381,1\n0.9339506644871263,1\n0.8784293563597917,1\n0.895670527899472,1\n0.8430529077285147,0\n0.9555244758421513,1\n0.6562334841941971,0\n0.44347198079416006,0\n0.5882951048748661,0\n0.6632882208074936,1\n0.7193471554059294,1\n0.4876146079345244,1\n0.953168038895694,0\n0.7174536271438119,0\n0.7623244598883582,1\n0.9509176055447596,1\n0.8831960622867566,1\n0.7383262444068864,1\n0.6705386587124225,1\n0.6742124297478073,0\n0.7473350702698447,1\n0.5874704389192519,1\n0.744815008138343,1\n0.8734280898226684,1\n0.8297216201264507,0\n0.500944448282882,1\n0.6082688643601383,1\n0.8503070978000848,1\n0.6915965862987412,1\n0.7453015147962215,1\n0.9149539864420778,0\n0.6881209119592449,0\n0.43798699977734107,1\n0.8914830297995384,0\n0.4105129789272475,0\n0.7472626135336042,1\n0.8677678395354927,0\n0.9018897593429304,1\n0.5551233908380523,1\n0.9424249586699289,1\n0.7490780290998588,0\n0.6472160899498884,1\n0.8677347123628618,1\n0.7553091985144603,1\n0.8320194462604729,0\n0.8075970235316438,0\n0.6339841866183518,1\n0.49726004536795126,0\n0.4105371902108986,1\n0.8417525930304077,1\n0.7536828171041157,1\n0.8064127928769447,1\n0.7854594038727644,1\n0.5963890082318981,0\n0.5529445130913578,0\n0.9267909229693727,1\n0.7515970304169407,1\n0.5593384722006013,0\n0.805210866278916,1\n0.7966683402613025,1\n0.5155310158049526,0\n0.4365883968646004,0\n0.7823467361944703,1\n0.7613503500913443,1\n0.626144209182107,1\n0.6694521261435745,1\n0.5866183003387513,1\n0.6000598269757904,0\n0.7172361276561078,1\n0.5527992496181331,0\n0.8590012221015682,1\n0.535433262307073,0\n0.8682061118295181,0\n0.9357722771399521,0\n0.8119736186824921,1\n0.7986123343128968,1\n0.8107896870003097,1\n0.8956489957173016,1\n0.7528890472068774,1\n0.7820916134883324,1\n0.9180589086435903,1\n0.7024068363271042,0\n0.7178566909297954,1\n0.7987040746674912,1\n0.8422369874987513,1\n0.7990381593486006,1\n0.734168410981497,1\n0.9148037540012764,0\n0.9512786598755222,1\n0.5115165194732113,0\n0.6680665924216662,0\n0.9035603892225401,1\n0.6278994800190909,1\n0.8197891964984755,0\n0.9647400144317287,0\n0.6972562919176344,1\n0.7684046697253971,0\n0.669647327062451,1\n0.7008342235525218,0\n0.5493029774934077,1\n0.829711646861401,0\n0.672083575156045,1\n0.8280829924700488,1\n0.7404069235501143,1\n0.9165172488447596,1\n0.2282370958780986,0\n0.5386581972300077,1\n0.7457243489462131,0\n0.5392385577908511,1\n0.7282403747702962,1\n0.8815371641998792,1\n0.9317419977093483,1\n0.6428026789281851,1\n0.430810850276538,0\n0.6506688477927165,1\n0.42472556381855225,0\n0.5521720366772187,0\n0.4962777211926458,0\n0.5521532981257298,0\n0.7796066865227138,0\n0.7969958630189081,0\n0.7912668283237764,1\n0.8779876427333928,1\n0.6233181788415615,1\n0.8342941272432719,1\n0.348248971986688,0\n0.836354674601855,0\n0.7486363732096073,1\n0.9022350531555352,1\n0.6235948805828943,0\n0.7412447353591232,1\n0.9513810859673139,1\n0.9228817311905471,1\n0.64916096179032,0\n0.9456648149743587,1\n0.5061343952388129,0\n0.8887209746062371,0\n0.79197745415576,0\n0.9066895906936023,0\n0.8363569154967893,1\n0.6975598059422804,0\n0.9110310794481414,1\n0.8845099497931554,1\n0.797249804568639,1\n0.8234974141447502,1\n0.8731182067678682,1\n0.6710415703971537,0\n0.6899353665103408,1\n0.9056899904434397,1\n0.5584019550584538,0\n0.6935218256066912,0\n0.5781141365894732,1\n0.5842441620215233,0\n0.8909719408973987,1\n0.47497656651502757,0\n0.30628057890831156,0\n0.6613257539649979,1\n0.9182098689670549,1\n0.8689256027144716,1\n0.7778303880886164,1\n0.6605842495341648,1\n0.7070577979094427,1\n0.5150620613284645,0\n0.7983389970158196,1\n0.44855480238672674,0\n0.6508648099224633,1\n0.9255033777884485,1\n0.394853735533111,0\n0.7867356781046114,1\n0.5862323242027035,1\n0.742574501218065,1\n0.8233157601980059,1\n0.9776754616159198,1\n0.8308139894160536,1\n0.6312823438927516,1\n0.7495943521641198,1\n0.6330588838243157,1\n0.805864860227732,0\n0.8887378106808479,1\n0.7164518065643302,1\n0.7346928331243497,1\n0.7964994861585364,0\n0.6452092959454038,1\n0.7887232244762341,1\n0.859688093786934,1\n0.7877313919164622,0\n0.8357431010713455,1\n0.761159160822158,1\n0.8012002285350062,0\n0.8342390553977881,0\n0.37358065987559713,0\n0.8399620235697624,0\n0.6093418214146425,1\n0.869359968523917,1\n0.7429978373115638,1\n0.9857316567739101,0\n0.8990737823380992,1\n0.7658433201169123,1\n0.8388790588565389,1\n0.13923361900684222,0\n0.7750684420668394,1\n0.5948658555396185,1\n0.4688512407957305,1\n0.9500019505154772,1\n0.6554731834099224,1\n0.7468502782246492,1\n0.6133090129809938,0\n0.6471700420621249,0\n0.9015034427957023,1\n0.918475057393014,1\n0.823568284769587,1\n0.8254780969183326,1\n0.8512121810403324,1\n0.7700545197165005,1\n0.715605522111546,1\n0.8236374691025015,0\n0.505662620472717,1\n0.7442785127318075,0\n0.5024241043670978,0\n0.5323628354510287,0\n0.49575715066601667,1\n0.8668514453238375,1\n0.6242494830337679,1\n0.7071417673637219,0\n0.46186034198055237,1\n0.8056259126568305,1\n0.3525147771800355,0\n0.842252928121914,1\n0.4873924680795274,1\n0.6706392345533406,0\n0.7386804567934326,1\n0.9086960236859598,1\n0.7764830732374907,1\n0.4095110358667931,1\n0.6654872908264021,1\n0.8415880166083428,1\n0.8703776906494938,0\n0.7233215805008749,0\n0.8981798649057454,1\n0.6446564490475859,1\n0.7307904831484313,1\n0.9337474280437011,1\n0.6230267811255795,1\n0.5360793418892547,1\n0.8660855378904423,1\n0.5271394905857312,0\n0.765759728799988,1\n0.6524696445069339,1\n0.9257168343443426,1\n0.5461358606182308,0\n0.7480755178612511,1\n0.8404996243858934,0\n0.46216000140207303,0\n0.939382629974081,1\n0.5308856836821055,1\n0.4263112730497274,1\n0.6318270693469151,0\n0.76546862303657,1\n0.8357280501480471,0\n0.710861165612637,1\n0.6217916985673697,0\n0.836028146464499,1\n0.8369846237927154,0\n0.8743917960956923,1\n0.8264623961107267,0\n0.8569061872208475,1\n", "n": 300, "ece": 0.08256759382915613, "mean": 0.05060766943179064, "p95": 0.07520325441503889, "statement": "ECE 0.0826 over 300 rows (10 equal width bins), against a calibrated-model floor of 0.0506 (95th percentile 0.0752): miscalibration is distinguishable from sampling noise." }, { "label": "tab separated, CRLF line ends, true and false, a blank line", "n_bins": 15, "text": "0.8092251254673093\tfalse\r\n0.4810475108402493\tfalse\r\n0.6150968308958921\ttrue\r\n0.16901035542560794\tfalse\r\n0.06109131264098788\tfalse\r\n0.12262722928878389\tfalse\r\n0.260205205042753\tfalse\r\n0.15746477725918107\tfalse\r\n0.09811156131315213\tfalse\r\n0.965616171921477\ttrue\r\n0.64024646917968\tfalse\r\n0.06283530440957164\tfalse\r\n0.6232950439881982\tfalse\r\n0.8174599894568946\ttrue\r\n0.4662516651391754\ttrue\r\n0.73528215473066\tfalse\r\n0.8560957030307116\ttrue\r\n0.7809314624426905\ttrue\r\n0.37469093722066327\tfalse\r\n0.8063052210769309\ttrue\r\n0.6797309842087644\tfalse\r\n0.5441444257254635\ttrue\r\n0.6640869675530805\ttrue\r\n0.8323301563131899\ttrue\r\n0.5994946338452651\ttrue\r\n0.3427956490611287\tfalse\r\n0.5574401539739183\ttrue\r\n0.936080333049103\ttrue\r\n0.6087827358989107\ttrue\r\n0.15745125025268203\tfalse\r\n0.6360353529084468\tfalse\r\n0.6914338569413809\ttrue\r\n0.42126265304268395\ttrue\r\n0.7833959783104201\ttrue\r\n0.8705602046249002\ttrue\r\n0.507211294929106\tfalse\r\n0.21790444828803646\tfalse\r\n0.6881032519474287\tfalse\r\n0.00411233163437319\tfalse\r\n0.3502617240759307\tfalse\r\n\r\n0.9352696943166048\ttrue\r\n0.6645994624576257\tfalse\r\n0.610127098529173\tfalse\r\n0.8208248356314631\ttrue\r\n0.92366253639984\ttrue\r\n0.6730912403808006\ttrue\r\n0.51129127814052\tfalse\r\n0.6817005717995038\ttrue\r\n0.8899861184206933\ttrue\r\n0.7140194045085091\tfalse\r\n0.9498130543981782\ttrue\r\n0.34520656461313304\tfalse\r\n0.21614137962877666\tfalse\r\n0.4783647743967546\ttrue\r\n0.8843747222028043\tfalse\r\n0.4061608268263315\tfalse\r\n0.0471967793994339\tfalse\r\n0.7220830451450202\ttrue\r\n0.7312581199981368\tfalse\r\n0.15257757557359208\tfalse\r\n0.6906773861917685\ttrue\r\n0.8159109793597631\tfalse\r\n0.9388029785643608\ttrue\r\n0.5196187225053817\ttrue\r\n0.005260178281515815\tfalse\r\n0.45596229708928215\tfalse\r\n0.6982282138608734\ttrue\r\n0.685603418704395\ttrue\r\n0.9719211064669567\ttrue\r\n0.15998676726080874\tfalse\r\n0.7850514335124563\ttrue\r\n0.5287084362267819\tfalse\r\n0.5118535986122918\tfalse\r\n0.3642922284946962\tfalse\r\n0.12612145219641568\tfalse\r\n0.8856591785934856\ttrue\r\n0.3500651481128494\tfalse\r\n0.8890678391009302\ttrue\r\n0.4778501255227805\tfalse\r\n0.2885141595192058\tfalse\r\n0.18314668398887002\tfalse\r\n0.8261608151809642\tfalse\r\n0.0670082083353638\tfalse\r\n0.699881437809391\tfalse\r\n0.000734218650418339\tfalse\r\n0.4763010765379868\ttrue\r\n0.7251867150355418\ttrue\r\n0.7643861262351269\ttrue\r\n0.3689060891729379\tfalse\r\n0.8407039905274829\ttrue\r\n0.1779477506507431\tfalse\r\n0.6525856250479789\ttrue\r\n0.11093291317605714\tfalse\r\n0.08169699773110173\tfalse\r\n0.8584770844072255\ttrue\r\n0.22227571522325595\tfalse\r\n0.37222805287748184\tfalse\r\n0.04131405978541758\tfalse\r\n0.18490454741039097\tfalse\r\n0.49893622010934036\tfalse\r\n0.5308800623040114\ttrue\r\n0.2124453901395038\tfalse\r\n0.8594821883376382\ttrue\r\n0.6141412143428835\ttrue\r\n0.049753841472731763\tfalse\r\n0.1270379908640027\tfalse\r\n0.3857686994077191\tfalse\r\n0.5290646649467213\ttrue\r\n0.23426514867427006\ttrue\r\n0.9721774387851962\ttrue\r\n0.8292813126608648\ttrue\r\n0.5643581363939282\tfalse\r\n0.7279323259932716\ttrue\r\n0.1733617511617992\tfalse\r\n0.5677147297527553\ttrue\r\n0.8417937724914958\ttrue\r\n0.7109214107389936\ttrue\r\n0.03928094042693042\ttrue\r\n0.5969846402370829\tfalse\r\n0.7096180600849119\ttrue\r\n0.06537837025450433\tfalse\r\n0.6958775599796297\ttrue\r\n0.1953942156617785\tfalse\r\n0.5557920253825632\ttrue\r\n0.9364734927925796\ttrue\r\n0.13308504047817282\tfalse\r\n0.9384531847054718\ttrue\r\n0.8059143590756168\ttrue\r\n0.1933381413865768\tfalse\r\n0.6408979420254313\ttrue\r\n0.09697128926361853\ttrue\r\n0.6062934627402057\tfalse\r\n0.31638374702764194\tfalse\r\n0.24657104375987926\tfalse\r\n0.05299896777931312\tfalse\r\n0.41718266278269156\tfalse\r\n0.9078211084000235\ttrue\r\n0.5898374349703307\tfalse\r\n0.2277324778059242\tfalse\r\n0.757064035744464\ttrue\r\n0.5688307708568273\tfalse\r\n0.00454309077941728\tfalse\r\n0.17939433257263737\tfalse\r\n0.08728479124377542\tfalse\r\n0.9373635851491866\ttrue\r\n0.21206081712181002\tfalse\r\n0.29868775220491417\tfalse\r\n0.6028032693583263\ttrue\r\n0.8043238663654723\tfalse\r\n0.9659887379506182\ttrue", "n": 150, "ece": 0.08763532023460592, "mean": 0.097457440489558, "p95": 0.13177332613847564, "statement": "ECE 0.0876 over 150 rows (15 equal width bins), against a calibrated-model floor of 0.0975 (95th percentile 0.1318): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable." }, { "label": "space separated, the ECE exactly on a rounding tie", "n_bins": 10, "text": " 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 1\n 0.75 0\n 0.75 0\n 0.75 0\n 0.75 0\n 0.75 0\n 0.75 0\n 0.75 0\n 0.75 0\n 0.75 0", "n": 32, "ece": 0.03125, "mean": 0.059703125, "p95": 0.15625, "statement": "ECE 0.0312 over 32 rows (10 equal width bins), against a calibrated-model floor of 0.0597 (95th percentile 0.1562): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable." }, { "label": "semicolons and quotes, the ECE beside a tie", "n_bins": 10, "text": "p;\"correct\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"1\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"\n\"0.80125\";\"0\"", "n": 400, "ece": 0.0012499999999998623, "mean": 0.016173749999999997, "p95": 0.03875000000000006, "statement": "ECE 0.0012 over 400 rows (10 equal width bins), against a calibrated-model floor of 0.0162 (95th percentile 0.0388): INCONCLUSIVE at this sample size. A perfectly calibrated model would often score this badly on this many rows, so this dataset cannot tell the two apart. This is not a clean bill of health: nothing was established either way. Collect more rows to make the question answerable." } ] }