OpenBench

OpenBench Testing Framework

Author	Menkib
Creation	1778258776
Source	https://github.com/Menkib64/lc0
Sha	5c4c0a3b0e30acae0810535c8fbd052818cb855e
Branch	uncertainty_weighting_logistic
Bench	1
Network	tf13-375
Options	Threads=2 Hash=1
Time	20.0+0.20
Opening Book	UHO_Lichess_4852_v1.epd
Upload PGNs	FALSE
Syzygy WDL	DISABLED
Syzygy ADJ.	OPTIONAL
Win ADJ.	movecount=3 score=400
Draw ADJ.	movenumber=40 movecount=8 score=10
Scaling	DEV 7000 NPS
Alpha	0.602
Gamma	0.101
A-Ratio	0.1
Reporting	BATCHED
Distribution	SINGLE
Pairs Per Point	16

uncertainty_weighting_lo... vs uncertainty_weighting_lo...

Approve Restart Delete

UncertaintyWeightingMidPoint, float, 0.04, 0.0, 1.0, 0.002, 0.002
UncertaintyWeightingMinusExponent, float, 80.0, 0.0, 100000.0, 0.8, 0.8
UncertaintyWeightingScale, float, 0.41, 0.0, 10.0, 0.02, 0.02
UncertaintyWeightingBase, float, 0.67, 0.0, 10.0, 0.02, 0.02
UncertaintyWeightingSkewRoot, float, 8.0, 0.01, 10000.0, 1.0, 1.0

Copy SPSA Inputs

UncertaintyWeightingMidPoint, 0.03991309695056231
UncertaintyWeightingMinusExponent, 106.02705397016564
UncertaintyWeightingScale, 0.40869929181134784
UncertaintyWeightingBase, 0.6566138880368153
UncertaintyWeightingSkewRoot, 4.744132127496989

Copy SPSA Outputs

Tuning 5 Parameters
119/25000 Iterations
3834/800000 Games Played

Name	Curr	Start	Min	Max	C	C_end	R	R_end
UncertaintyWeightingMidPoint	0.0399	0.0400	0.0000	1.0000	0.0034	0.0020	0.0028	0.0020
UncertaintyWeightingMinusExponent	106.0271	80.0000	0.0000	100000.0000	1.3708	0.8000	1.1217	0.8000
UncertaintyWeightingScale	0.4087	0.4100	0.0000	10.0000	0.0343	0.0200	0.0280	0.0200
UncertaintyWeightingBase	0.6566	0.6700	0.0000	10.0000	0.0343	0.0200	0.0280	0.0200
UncertaintyWeightingSkewRoot	4.7441	8.0000	0.0100	10000.0000	1.7136	1.0000	1.4021	1.0000

#	User	Updated	Games	Won	Lost	Drawn
1253	Kovax	1778263626	86	19	19	48
239	Ploot	1778263660	72	17	16	39
413	Kovax	1778263662	72	19	15	38
1265	Kovax	1778263660	86	24	27	35
1309	Kovax	1778263632	82	25	21	36
1306	Kovax	1778263652	84	24	21	39
1282	Kovax	1778263667	86	19	19	48
1311	Kovax	1778263600	86	28	21	37
1274	Kovax	1778263638	78	18	17	43
1281	Kovax	1778263572	84	17	17	50
236	Ploot	1778263543	68	12	16	40
1261	Kovax	1778263600	84	14	15	55
238	Ploot	1778263563	68	12	17	39
1258	Kovax	1778263626	82	24	23	35
1271	Kovax	1778263551	84	21	18	45
1307	Kovax	1778263561	84	18	19	47
1263	Kovax	1778263643	88	23	26	39
1257	Kovax	1778263607	84	19	20	45
1308	Kovax	1778263597	86	21	20	45
1269	Kovax	1778263576	82	22	19	41
1275	Kovax	1778263633	80	19	15	46
1313	Kovax	1778263552	82	22	17	43
1280	Kovax	1778263670	86	25	22	39
1312	Kovax	1778263567	86	24	28	34
1267	Kovax	1778263581	82	17	23	42
416	Kovax	1778263594	66	16	13	37
1279	Kovax	1778263587	88	27	25	36
1266	Kovax	1778263624	88	26	25	37
1256	Kovax	1778263592	86	23	22	41
1268	Kovax	1778263612	80	15	14	51
1310	Kovax	1778263634	84	20	22	42
1278	Kovax	1778263576	86	23	26	37
1273	Kovax	1778263656	86	19	19	48
1262	Kovax	1778263588	82	20	24	38
415	Kovax	1778263618	70	22	19	29
1255	Kovax	1778263611	86	22	18	46
1259	Kovax	1778263566	82	23	24	35
1276	Kovax	1778263658	86	22	22	42
1252	Kovax	1778263620	86	25	23	38
1270	Kovax	1778263616	82	18	18	46
1272	Kovax	1778263604	80	17	21	42
1264	Kovax	1778263597	82	23	20	39
237	Ploot	1778263564	70	20	19	31
414	Kovax	1778263607	66	15	18	33
1277	Kovax	1778263628	86	23	23	40
1260	Kovax	1778263664	84	14	19	51
1283	Kovax	1778263618	86	26	27	33