learning_rate: 1e-05
train_batch_size: 32
eval_batch_size: 8
seed: 42
optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.4274	0.0243	5	1.2228	95.9414
1.227	0.0485	10	1.1423	98.5626
0.9722	0.0728	15	1.0273	107.6099
0.9451	0.0971	20	0.9282	110.6821
0.7808	0.1214	25	0.8426	106.9899
0.8525	0.1456	30	0.7748	95.3495
0.7257	0.1699	35	0.7515	102.5085
0.6942	0.1942	40	0.7251	99.9154
0.5966	0.2184	45	0.7008	99.8309
0.6253	0.2427	50	0.6919	98.9572
0.7938	0.2670	55	0.6732	79.2841
0.5321	0.2913	60	0.6781	91.2909
0.5656	0.3155	65	0.6678	77.2266
0.5561	0.3398	70	0.6609	89.6843
0.5397	0.3641	75	0.6507	88.3315
0.5613	0.3883	80	0.6455	80.1297
0.4692	0.4126	85	0.6458	80.2142
0.5611	0.4369	90	0.6380	76.9729
0.4138	0.4612	95	0.6572	93.0665
0.6113	0.4854	100	0.6382	74.8873
0.5782	0.5097	105	0.6517	85.4848
0.5665	0.5340	110	0.6340	82.1871
0.6359	0.5583	115	0.6346	80.4397
0.3424	0.5825	120	0.6306	76.9166
0.3428	0.6068	125	0.6521	84.3574
0.4731	0.6311	130	0.6341	78.9741
0.3675	0.6553	135	0.6324	78.3258
0.4426	0.6796	140	0.6422	87.1759
0.4387	0.7039	145	0.6243	78.5795
0.3983	0.7282	150	0.6232	79.7632
0.3755	0.7524	155	0.6381	86.9222
0.3717	0.7767	160	0.6340	76.6911
0.3379	0.8010	165	0.6499	92.1082
0.4277	0.8252	170	0.6249	79.7069
0.4051	0.8495	175	0.6184	84.3856
0.359	0.8738	180	0.6170	81.2852
0.2938	0.8981	185	0.6303	81.9053
0.3446	0.9223	190	0.6316	81.9617
0.4052	0.9466	195	0.6407	79.2277
0.3562	0.9709	200	0.6290	77.4521
0.2911	0.9951	205	0.6315	86.1048
0.1966	1.0194	210	0.6436	81.7644
0.2656	1.0437	215	0.6562	82.2435
0.1923	1.0680	220	0.6587	82.1308
0.2252	1.0922	225	0.6556	83.4273
0.2853	1.1165	230	0.6516	84.9493
0.2731	1.1408	235	0.6546	85.5693
0.2119	1.1650	240	0.6462	84.2728
0.2352	1.1893	245	0.6564	86.0767
0.2825	1.2136	250	0.6591	87.7114
0.2755	1.2379	255	0.6477	79.3123
0.197	1.2621	260	0.6513	80.7497
0.2387	1.2864	265	0.6421	80.1578
0.1785	1.3107	270	0.6700	82.1590
0.1771	1.3350	275	0.6633	77.5366
0.2325	1.3592	280	0.6855	82.8636
0.1546	1.3835	285	0.6711	83.3427
0.2431	1.4078	290	0.6610	78.9459
0.2418	1.4320	295	0.6604	80.7497
0.2455	1.4563	300	0.6631	78.5513
0.1826	1.4806	305	0.6676	84.3010
0.1851	1.5049	310	0.6561	81.1443
0.215	1.5291	315	0.6729	78.6359
0.1917	1.5534	320	0.6660	77.8749
0.1268	1.5777	325	0.6589	81.9053
0.2435	1.6019	330	0.6680	86.3021
0.1929	1.6262	335	0.6826	91.7136
0.131	1.6505	340	0.6669	85.6539
0.1832	1.6748	345	0.6581	77.3112
0.136	1.6990	350	0.6645	77.5366
0.1904	1.7233	355	0.6677	80.5524
0.1749	1.7476	360	0.6811	77.0293
0.1647	1.7718	365	0.6802	79.5096
0.1748	1.7961	370	0.6600	78.1285
0.1653	1.8204	375	0.6657	80.5242
0.2109	1.8447	380	0.6640	82.0180
0.2229	1.8689	385	0.6873	79.5660
0.2415	1.8932	390	0.6860	78.2131
0.1555	1.9175	395	0.6855	85.9357
0.1799	1.9417	400	0.6793	84.4983
0.1647	1.9660	405	0.6748	77.1139
0.1899	1.9903	410	0.6919	83.9064
0.0603	2.0146	415	0.6913	79.1432
0.0933	2.0388	420	0.7156	78.1849
0.0886	2.0631	425	0.7352	80.2142
0.0918	2.0874	430	0.7267	77.1421
0.0873	2.1117	435	0.7305	84.8647
0.0633	2.1359	440	0.7306	81.6798
0.097	2.1602	445	0.7405	81.8489
0.1064	2.1845	450	0.7450	80.3551
0.1034	2.2087	455	0.7436	84.1319
0.0888	2.2330	460	0.7430	84.3574
0.0829	2.2573	465	0.7244	79.5378
0.0496	2.2816	470	0.7371	86.2176
0.0895	2.3058	475	0.7235	80.4115
0.0391	2.3301	480	0.7266	75.7328
0.0895	2.3544	485	0.7453	84.4419
0.1079	2.3786	490	0.7393	76.8320
0.0445	2.4029	495	0.7391	80.7215
0.0814	2.4272	500	0.7488	86.7813
0.1023	2.4515	505	0.7224	83.1736
0.0874	2.4757	510	0.7231	79.4532
0.1106	2.5	515	0.7344	80.9752
0.0491	2.5243	520	0.7425	81.2007
0.0724	2.5485	525	0.7570	81.8489
0.0678	2.5728	530	0.7599	80.5806
0.0827	2.5971	535	0.7463	81.6234
0.0593	2.6214	540	0.7326	77.3393
0.06	2.6456	545	0.7532	81.9335
0.0733	2.6699	550	0.7579	80.8343
0.0387	2.6942	555	0.7532	79.8196
0.037	2.7184	560	0.7532	80.4397
0.0363	2.7427	565	0.7477	78.4104
0.0608	2.7670	570	0.7548	78.9741
0.0544	2.7913	575	0.7551	81.2007
0.1091	2.8155	580	0.7595	77.1702
0.0771	2.8398	585	0.7713	79.7351
0.0694	2.8641	590	0.7713	77.7621
0.041	2.8883	595	0.7743	78.2131
0.0925	2.9126	600	0.7729	78.1285
0.0568	2.9369	605	0.7607	76.6347
0.0763	2.9612	610	0.7595	75.9865
0.072	2.9854	615	0.7633	77.8467
0.0118	3.0097	620	0.7678	79.3968
0.0568	3.0340	625	0.7815	78.2413
0.0251	3.0583	630	0.8020	77.7621
0.038	3.0825	635	0.8161	77.0857
0.0447	3.1068	640	0.8272	77.9594
0.03	3.1311	645	0.8253	76.8320
0.0384	3.1553	650	0.8170	75.9301
0.0278	3.1796	655	0.8157	76.8320
0.0187	3.2039	660	0.8246	79.6223
0.0411	3.2282	665	0.8221	79.2277
0.0507	3.2524	670	0.8210	78.6077
0.0476	3.2767	675	0.8186	77.9594
0.0193	3.3010	680	0.8237	78.5795
0.0333	3.3252	685	0.8295	78.9741
0.0182	3.3495	690	0.8296	77.7621
0.0399	3.3738	695	0.8336	78.2413
0.023	3.3981	700	0.8299	76.5220
0.0351	3.4223	705	0.8268	76.5502
0.0521	3.4466	710	0.8286	77.5648
0.0127	3.4709	715	0.8274	78.2694
0.0304	3.4951	720	0.8183	76.1838
0.0577	3.5194	725	0.8164	74.6900
0.0262	3.5437	730	0.8254	75.7046
0.0154	3.5680	735	0.8297	76.6911
0.0492	3.5922	740	0.8273	76.6629
0.0169	3.6165	745	0.8310	77.4803
0.0338	3.6408	750	0.8355	78.4667
0.0336	3.6650	755	0.8327	77.7903
0.029	3.6893	760	0.8275	76.0428
0.0265	3.7136	765	0.8247	75.9865
0.0294	3.7379	770	0.8294	77.4521
0.0315	3.7621	775	0.8372	79.1714
0.0504	3.7864	780	0.8349	79.1150
0.0151	3.8107	785	0.8320	78.1567
0.0137	3.8350	790	0.8302	77.0857
0.0321	3.8592	795	0.8282	76.4656
0.0245	3.8835	800	0.8281	76.4374
0.0191	3.9078	805	0.8297	76.9166
0.0225	3.9320	810	0.8312	77.6494
0.0255	3.9563	815	0.8322	77.4521
0.0195	3.9806	820	0.8331	77.6212

Framework versions

Transformers 4.57.3
Pytorch 2.9.1+cu128
Datasets 4.4.1
Tokenizers 0.22.1

nkkbr/whisper-large-v3-zatoichi-ja-EX-9-combine

Whisper Large v3 - Japanese Zatoichi ASR

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

nkkbr/
whisper-large-v3-zatoichi-ja-EX-9-combine