CHON

as donto believed on

From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)

CHON has 68 facts recorded in Dontopedia across 6 references, with 9 live disagreements.

68 facts·51 predicates·6 sources·9 in dispute

Mostly:has training time(5), has recall(3), has parameter count(3)

Inbound mentions (7)

Other subjects in dontopedia point AT this entity as a value. These are inverse relationships — e.g. "X motherOf this subject" — and answer questions the forward facts can't. Grouped by predicate.

`advocatesFor`Advocates for(1)

Xenonfunex:xenonfun

`affectsPerformanceOf`Affects Performance of(1)

Expanding Fault Windowex:expanding-fault-window

`areAlreadyGood`Are Already Good(1)

Early Life Predictionsex:early-life-predictions

`assertsModelBehavior`Asserts Model Behavior(1)

Xenonfunex:xenonfun

`comparesModels`Compares Models(1)

Formal Results Tableex:formal-results-table

`partOf`Part of(1)

Ema Detectorex:ema-detector

`usedForBenchmarking`Used for Benchmarking(1)

NASA Ims Bearing Datasetex:nasa-ims-bearing-dataset

Other facts (66)

The long tail: predicates that appear too rarely to warrant their own section. Filter or scroll to find a specific one. Each row links to its source.

66 facts

Predicate	Value	Ref
Has Training Time	0.1 second	[1]
Has Training Time	0.1	[2]
Has Training Time	0.1 sec	[2]
Has Training Time	0.1 second	[5]
Has Training Time	0.1	[6]
Has Recall	100%	[1]
Has Recall	1	[2]
Has Recall	100	[6]
Has Parameter Count	119	[2]
Has Parameter Count	116	[5]
Has Parameter Count	119	[6]
Has Accuracy	0.902	[2]
Has Accuracy	90.2%	[6]
Has Precision	0.676	[2]
Has Precision	67.6	[6]
Has Auroc Score	0.985	[2]
Has Auroc Score	0.98	[2]
Has F1 Score	0.81	[2]
Has F1 Score	0.81	[6]
Has Fewer Parameters Than	Best Published Lstm	[2]
Has Fewer Parameters Than	Lstm Autoencoder	[2]
Detects Earlier Than	Ground Truth Fault Window	[2]
Detects Earlier Than	Binary Label Fault Start	[5]
Rdf:type	Model	[4]
Rdf:type	Machine Learning Model	[6]
Achieves Recall	100%	[1]
Achieves Auroc for Fault25 Percent	0.989	[1]
Achieves Auroc for Fault30 Percent	0.98	[1]
Achieves F1	0.826	[1]
Achieves F1 for Fault25 Percent	0.932	[1]
Achieves F1 for Fault30 Percent	0.993	[1]
Achieves Near Perfect F1	null	[1]
Achieves Precision	70.4%	[1]
Achieves Precision for Fault25 Percent	87.3%	[1]
Achieves Precision for Fault30 Percent	98.7%	[1]
Achieves Recall for Fault25 Percent	100%	[1]
Achieves Recall for Fault30 Percent	100%	[1]
Beats Every	Published Baseline	[1]
Beats on F1	Lstm Autoencoder	[1]
Achieves Auroc	0.985	[1]
Detects Degradation	Earlier Than Binary Label	[1]
Superior to Baselines	null	[1]
Detects Early Degradation	null	[1]
Has860 Times Fewer Parameters Than	Lstm Autoencoder	[1]
Has Auroc	0.98	[1]
Has F1	0.993	[1]
Has Number of Parameters	116	[1]
Is Not Making	False Positives	[1]
Evaluated on	NASA Ims Bearing Dataset	[2]
Outperforms in Training Speed	Lstm Autoencoder	[2]
Matches Auroc of	Lstm Autoencoder	[2]
Is Ours	null	[2]
Claimed Superior in Efficiency	null	[2]
Has False Positives Count	24	[2]
Evaluated on Full Results	null	[3]
Test Status	unchanged	[4]
Has F1 Score at Fault30pct	0.993	[5]
Has Precision at Fault30pct	98.7%	[5]
Has Recall at Fault30pct	100%	[5]
Has Auroc at Fault30pct	0.98	[5]
Detects	Degradation	[5]
Outperforms	Published Baselines	[5]
Has Parameter Count Advantage Over	Lstm Autoencoder	[5]
Has Auroc	0.985	[6]
Measured in Dataset	NASA Ims Bearing Dataset	[6]
Has Owner	Our Team	[6]

Timeline axis is valid_time — when each source says the fact was true in the world, not when Dontopedia learned about it. Retracted rows are kept for provenance; coloured stripes indicate the context kind.

68 undated

—

achievesRecallblah/watt-activation/part-509

100%

—

achievesAurocForFault25Percentblah/watt-activation/part-509

0.989

—

achievesAurocForFault30Percentblah/watt-activation/part-509

0.98

—

achievesF1blah/watt-activation/part-509

0.826

—

achievesF1ForFault25Percentblah/watt-activation/part-509

0.932

—

achievesF1ForFault30Percentblah/watt-activation/part-509

0.993

—

achievesNearPerfectF1blah/watt-activation/part-509

null

—

achievesPrecisionblah/watt-activation/part-509

70.4%

—

achievesPrecisionForFault25Percentblah/watt-activation/part-509

87.3%

—

achievesPrecisionForFault30Percentblah/watt-activation/part-509

98.7%

—

achievesRecallForFault25Percentblah/watt-activation/part-509

100%

—

achievesRecallForFault30Percentblah/watt-activation/part-509

100%

—

beatsEveryblah/watt-activation/part-509

ex:published-baseline

—

beatsOnF1blah/watt-activation/part-509

ex:lstm-autoencoder

—

achievesAurocblah/watt-activation/part-509

0.985

—

detectsDegradationblah/watt-activation/part-509

ex:earlier-than-binary-label

—

superiorToBaselinesblah/watt-activation/part-509

null

—

detectsEarlyDegradationblah/watt-activation/part-509

null

—

has860TimesFewerParametersThanblah/watt-activation/part-509

ex:lstm-autoencoder

—

hasAurocblah/watt-activation/part-509

0.98

—

hasF1blah/watt-activation/part-509

0.993

—

hasNumberOfParametersblah/watt-activation/part-509

116

—

isNotMakingblah/watt-activation/part-509

ex:false-positives

—

hasTrainingTimeblah/watt-activation/part-509

0.1 second

—

hasRecallblah/watt-activation/part-509

100%

—

hasAccuracyblah/watt-activation/part-508

0.902

—

hasParameterCountblah/watt-activation/part-508

119

—

hasPrecisionblah/watt-activation/part-508

0.676

—

evaluatedOnblah/watt-activation/part-508

ex:nasa-ims-bearing-dataset

—

hasTrainingTimeblah/watt-activation/part-508

0.1

—

outperformsInTrainingSpeedblah/watt-activation/part-508

ex:lstm-autoencoder

—

hasAurocScoreblah/watt-activation/part-508

0.985

—

hasAurocScoreblah/watt-activation/part-508

0.98

—

hasRecallblah/watt-activation/part-508

—

hasTrainingTimeblah/watt-activation/part-508

0.1 sec

—

hasF1Scoreblah/watt-activation/part-508

0.81

—

matchesAurocOfblah/watt-activation/part-508

ex:lstm-autoencoder

—

isOursblah/watt-activation/part-508

null

—

claimedSuperiorInEfficiencyblah/watt-activation/part-508

null

—

hasFalsePositivesCountblah/watt-activation/part-508

—

hasFewerParametersThanblah/watt-activation/part-508

ex:best-published-lstm

—

hasFewerParametersThanblah/watt-activation/part-508

ex:lstm-autoencoder

—

detectsEarlierThanblah/watt-activation/part-508

ex:ground-truth-fault-window

—

evaluatedOnFullResultsblah/watt-activation/part-515

null

—

typeblah/watt-activation/501

ex:Model

—

labelblah/watt-activation/501

CHON

—

testStatusblah/watt-activation/501

unchanged

—

hasF1ScoreAtFault30pctblah/watt-activation/506

0.993

—

hasPrecisionAtFault30pctblah/watt-activation/506

98.7%

—

hasRecallAtFault30pctblah/watt-activation/506

100%

—

hasAurocAtFault30pctblah/watt-activation/506

0.98

—

hasParameterCountblah/watt-activation/506

116

—

detectsblah/watt-activation/506

ex:degradation

—

detectsEarlierThanblah/watt-activation/506

ex:binary-label-fault-start

—

outperformsblah/watt-activation/506

ex:published-baselines

—

hasTrainingTimeblah/watt-activation/506

0.1 second

—

hasParameterCountAdvantageOverblah/watt-activation/506

ex:lstm-autoencoder

—

typeblah/watt-activation/505

ex:MachineLearningModel

—

labelblah/watt-activation/505

CHON

—

hasParameterCountblah/watt-activation/505

119

—

hasF1Scoreblah/watt-activation/505

0.81

—

hasAUROCblah/watt-activation/505

0.985

—

hasAccuracyblah/watt-activation/505

90.2%

—

hasRecallblah/watt-activation/505

100

—

hasPrecisionblah/watt-activation/505

67.6

—

hasTrainingTimeblah/watt-activation/505

0.1

—

measuredInDatasetblah/watt-activation/505

ex:nasa-ims-bearing-dataset

—

hasOwnerblah/watt-activation/505

ex:our-team

References (6)

6 references

[1]Part 50925 facts
ctx:discord/blah/watt-activation/part-509
[2]Part 50818 facts
ctx:discord/blah/watt-activation/part-508
[3]Part 5151 fact
ctx:discord/blah/watt-activation/part-515
[4]5013 facts
ctx:discord/blah/watt-activation/501
- full textwatt-activation-501
  text/plain2 KBdoc:agent/watt-activation-501/185a146d-ca90-4ea8-8412-733781e20533
  Show excerpt
  [2026-03-22 19:35] xenonfun: ``` --- What We're NOT Porting (and Why) - Spatial Clifford convolutions (1D/2D/3D): We do sequence models, not spatial CNNs. Fourier convolution covers the spectral case. - PR #17 grouped conv fix: Only ap…
[5]50610 facts
ctx:discord/blah/watt-activation/506
- full textwatt-activation-506
  text/plain2 KBdoc:agent/watt-activation-506/491a6927-81ff-44b4-967f-83424796e025
  Show excerpt
  [2026-03-22 20:23] xenonfun: ⏺ Interesting — AUROC is identical at 0.985 across all group counts. The residual separability is already maxed out. The F1 differences come from threshold tuning, not the model itself. G=1 actually gives the …
[6]50511 facts
ctx:discord/blah/watt-activation/505
- full textwatt-activation-505
  text/plain2 KBdoc:agent/watt-activation-505/942386e0-0cb8-490c-8592-6512fb5cc7cb
  Show excerpt
  [2026-03-22 20:11] xenonfun: ``` ⏺ Formal results: ┌───────────────┬───────────────────┬──────────────────────────┐ │ Metric │ CHON (119 params) │ Best Published │ ├───────────────┼───────────────────┼───────────────…

Keep researching

Missing something or suspicious of what's here? Kick off a research session — a Claude agent will investigate, cite its sources, and file new facts into a dedicated context you can review before accepting into the shared view.

CHON

Inbound mentions (7)

advocatesForAdvocates for(1)

affectsPerformanceOfAffects Performance of(1)

areAlreadyGoodAre Already Good(1)

assertsModelBehaviorAsserts Model Behavior(1)

comparesModelsCompares Models(1)

partOfPart of(1)

usedForBenchmarkingUsed for Benchmarking(1)

Other facts (66)

Timeline

References (6)

See also

Keep researching

`advocatesFor`Advocates for(1)

`affectsPerformanceOf`Affects Performance of(1)

`areAlreadyGood`Are Already Good(1)

`assertsModelBehavior`Asserts Model Behavior(1)

`comparesModels`Compares Models(1)

`partOf`Part of(1)

`usedForBenchmarking`Used for Benchmarking(1)