learning_rate: 0.001
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Use OptimizerNames.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_steps: 100
num_epochs: 30
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer	Wer
0.7249	0.2764	200	0.5182	0.1882	0.8634
0.5133	0.5529	400	0.4100	0.1605	0.7678
0.4887	0.8293	600	0.4116	0.1551	0.7577
0.4455	1.1050	800	0.3681	0.1459	0.7295
0.4346	1.3815	1000	0.3604	0.1442	0.7101
0.4326	1.6579	1200	0.3446	0.1426	0.6977
0.4236	1.9343	1400	0.3420	0.1430	0.7037
0.3989	2.2101	1600	0.3619	0.1448	0.7061
0.3903	2.4865	1800	0.3352	0.1404	0.6895
0.3854	2.7630	2000	0.3294	0.1375	0.6817
0.3855	3.0387	2200	0.3351	0.1430	0.7034
0.3837	3.3151	2400	0.3187	0.1383	0.6809
0.3709	3.5916	2600	0.3161	0.1341	0.6613
0.3852	3.8680	2800	0.3239	0.1350	0.6778
0.3691	4.1437	3000	0.3216	0.1384	0.6797
0.3585	4.4202	3200	0.3199	0.1357	0.6683
0.3719	4.6966	3400	0.3243	0.1364	0.6647
0.4083	4.9730	3600	0.3746	0.1457	0.7017
0.4498	5.2488	3800	0.3717	0.1384	0.6793
0.5022	5.5252	4000	0.4116	0.1413	0.7046
0.5274	5.8017	4200	0.4124	0.1477	0.7106
0.6096	6.0774	4400	0.5763	0.1553	0.7658
2.4044	6.3538	4600	2.4099	0.6829	1.0555
2.6559	6.6303	4800	2.4072	0.6403	1.0703
2.4745	6.9067	5000	2.2001	0.5914	1.1017
2.2546	7.1824	5200	2.0233	0.4348	1.1306
2.1064	7.4589	5400	1.8846	0.4351	1.1789
2.0695	7.7353	5600	1.8752	0.4194	1.1315
2.0974	8.0111	5800	1.9403	0.3747	1.0883
2.1528	8.2875	6000	1.9997	0.3921	1.0938
2.2246	8.5639	6200	2.0428	0.3714	1.0878
2.2192	8.8404	6400	2.1733	0.3318	1.0703
3.4126	9.1161	6600	3.7412	0.3576	1.1062
5.4137	9.3925	6800	5.4672	0.4165	1.0921
6.1068	9.6690	7000	6.1162	0.4302	1.1081
7.2401	9.9454	7200	6.8656	0.4661	1.1459
7.2621	10.2211	7400	6.8046	0.4855	1.2167
7.6861	10.4976	7600	7.4875	0.6845	1.2829
7.4586	10.7740	7800	7.2436	0.6340	1.1789
7.4007	11.0498	8000	7.2299	0.6331	1.1791
7.4492	11.3262	8200	7.2306	0.6336	1.1806
7.7305	11.6026	8400	7.2306	0.6342	1.1799
7.1362	11.8791	8600	7.2299	0.6335	1.1799
7.3227	12.1548	8800	7.2302	0.6331	1.1796
7.4756	12.4312	9000	7.2292	0.6329	1.1784
7.2954	12.7077	9200	7.2301	0.6334	1.1794
7.4613	12.9841	9400	7.2303	0.6332	1.1794
7.3689	13.2598	9600	7.2293	0.6339	1.1791
7.4248	13.5363	9800	7.2306	0.6334	1.1808
7.6585	13.8127	10000	7.2299	0.6342	1.1794
7.2369	14.0885	10200	7.2297	0.6332	1.1799
7.3086	14.3649	10400	7.2303	0.6332	1.1808
7.3882	14.6413	10600	7.2311	0.6341	1.1808
7.396	14.9178	10800	7.2293	0.6333	1.1796
7.4106	15.1935	11000	7.2305	0.6339	1.1803
7.4864	15.4699	11200	7.2306	0.6335	1.1787
7.2362	15.7464	11400	7.2295	0.6334	1.1789
7.3786	16.0221	11600	7.2304	0.6337	1.1789
7.3678	16.2985	11800	7.2307	0.6334	1.1794
7.3747	16.5750	12000	7.2300	0.6340	1.1801
7.4581	16.8514	12200	7.2301	0.6333	1.1803
7.3916	17.1272	12400	7.2304	0.6342	1.1803
6.9549	17.4036	12600	7.2300	0.6333	1.1801
7.5896	17.6800	12800	7.2300	0.6334	1.1803
7.3437	17.9565	13000	7.2293	0.6332	1.1796
7.1659	18.2322	13200	7.2299	0.6336	1.1787
7.3521	18.5086	13400	7.2299	0.6335	1.1796
7.388	18.7851	13600	7.2302	0.6339	1.1801
7.6184	19.0608	13800	7.2305	0.6335	1.1782
7.3446	19.3372	14000	7.2299	0.6333	1.1789
7.5834	19.6137	14200	7.2300	0.6339	1.1799
7.0571	19.8901	14400	7.2300	0.6331	1.1789
7.5434	20.1659	14600	7.2301	0.6335	1.1789
7.5061	20.4423	14800	7.2297	0.6334	1.1777
7.3337	20.7187	15000	7.2301	0.6334	1.1794
7.5379	20.9952	15200	7.2293	0.6337	1.1789
7.3616	21.2709	15400	7.2300	0.6338	1.1808
7.4563	21.5473	15600	7.2301	0.6342	1.1803
7.5492	21.8238	15800	7.2298	0.6338	1.1782
7.1092	22.0995	16000	7.2296	0.6340	1.1801
7.5108	22.3760	16200	7.2304	0.6327	1.1801
7.3233	22.6524	16400	7.2296	0.6329	1.1796
7.13	22.9288	16600	7.2309	0.6335	1.1794
7.2863	23.2046	16800	7.2290	0.6332	1.1787
7.2786	23.4810	17000	7.2298	0.6332	1.1782
7.3578	23.7574	17200	7.2300	0.6339	1.1803
7.3185	24.0332	17400	7.2295	0.6332	1.1789
7.7091	24.3096	17600	7.2309	0.6339	1.1808
7.305	24.5860	17800	7.2302	0.6336	1.1801
7.4239	24.8625	18000	7.2300	0.6339	1.1791
7.1861	25.1382	18200	7.2297	0.6334	1.1801
7.5163	25.4147	18400	7.2294	0.6334	1.1787
7.6509	25.6911	18600	7.2302	0.6336	1.1796
7.4563	25.9675	18800	7.2299	0.6332	1.1794
7.3944	26.2433	19000	7.2296	0.6337	1.1799
7.5816	26.5197	19200	7.2297	0.6336	1.1791
7.181	26.7961	19400	7.2299	0.6336	1.1794
7.4329	27.0719	19600	7.2297	0.6337	1.1801
7.4453	27.3483	19800	7.2304	0.6340	1.1799
7.3638	27.6247	20000	7.2305	0.6339	1.1791
7.3577	27.9012	20200	7.2307	0.6334	1.1808
7.3986	28.1769	20400	7.2304	0.6337	1.1789
7.3389	28.4534	20600	7.2300	0.6339	1.1791
7.2829	28.7298	20800	7.2302	0.6327	1.1787
7.5371	29.0055	21000	7.2294	0.6334	1.1796
7.3099	29.2820	21200	7.2298	0.6335	1.1801
7.3265	29.5584	21400	7.2297	0.6330	1.1808
7.3291	29.8348	21600	7.2292	0.6341	1.1794

Framework versions

Transformers 4.57.2
Pytorch 2.9.1+cu128
Datasets 3.6.0
Tokenizers 0.22.0

ctaguchi/ssc-ady-mms-model-mix-adapt-max

ssc-ady-mms-model-mix-adapt-max

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

ctaguchi/
ssc-ady-mms-model-mix-adapt-max