learning_rate: 0.0003
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
7.5593	0.1705	100	3.7574	1.0	1.0
3.7617	0.3410	200	3.6406	1.0	1.0
3.7935	0.5115	300	3.7703	1.0000	1.0
3.7515	0.6820	400	3.4756	0.9938	0.9993
3.7554	0.8525	500	3.5054	0.9932	0.9998
3.7684	1.0222	600	3.4862	0.9932	1.0
3.7297	1.1927	700	3.5204	0.9933	1.0
3.7861	1.3632	800	3.6136	0.9932	1.0
3.7326	1.5337	900	3.6448	0.9932	1.0
3.7333	1.7042	1000	3.5650	0.9932	1.0
3.7813	1.8747	1100	3.5611	0.9932	0.9998
3.6991	2.0443	1200	4.0765	0.9932	1.0
3.783	2.2148	1300	3.8187	0.9932	1.0
3.812	2.3853	1400	3.8855	0.9932	1.0
3.7223	2.5558	1500	3.8915	0.9932	0.9998
3.7504	2.7263	1600	3.8943	0.9932	1.0
3.7417	2.8968	1700	3.9197	0.9930	1.0
3.7336	3.0665	1800	3.5882	0.9938	1.0
3.7272	3.2370	1900	3.5592	0.9938	1.0
3.7303	3.4075	2000	3.5324	0.9856	1.0
3.7643	3.5780	2100	3.5002	0.9847	1.0
3.7024	3.7485	2200	3.4924	0.9931	0.9998
3.7144	3.9190	2300	3.5628	0.9932	1.0
3.6976	4.0887	2400	3.4476	0.9930	1.0
3.7041	4.2592	2500	3.4516	0.9932	1.0
3.708	4.4297	2600	3.4573	0.9932	1.0
3.6759	4.6002	2700	3.4665	0.9938	1.0
3.668	4.7707	2800	3.4816	0.9932	1.0
3.6309	4.9412	2900	3.4814	0.9930	1.0
3.6628	5.1108	3000	4.2411	0.9938	1.0
3.6599	5.2813	3100	4.2083	0.9932	0.9998
3.5753	5.4518	3200	4.0952	0.9890	1.0
3.6248	5.6223	3300	4.5791	0.9938	1.0
3.5746	5.7928	3400	4.4709	0.9939	1.0
3.5916	5.9633	3500	4.0641	0.9864	0.9986
3.5661	6.1330	3600	4.0738	0.9938	1.0
3.5995	6.3035	3700	3.8767	0.9863	1.0
3.5785	6.4740	3800	4.2056	0.9864	0.9984
3.6001	6.6445	3900	4.2901	0.9932	1.0
3.5385	6.8150	4000	4.2071	0.9819	1.0
3.5545	6.9855	4100	4.0290	0.9903	0.9991
3.5021	7.1552	4200	3.6264	0.9868	0.9984
3.5614	7.3257	4300	3.6253	0.9932	1.0
3.5401	7.4962	4400	4.2398	0.9877	0.9998
3.5546	7.6667	4500	3.5696	0.9866	1.0
3.5081	7.8372	4600	3.6183	0.9776	1.0
3.5486	8.0068	4700	3.8877	0.9809	1.0
3.4921	8.1773	4800	3.9502	0.9809	1.0
3.4263	8.3478	4900	3.8585	0.9876	0.9998
3.4313	8.5183	5000	4.0835	0.9759	0.9991
3.4208	8.6888	5100	3.7700	0.9689	1.0
3.392	8.8593	5200	3.8942	0.9761	1.0
3.3797	9.0290	5300	3.7086	0.9662	1.0
3.3889	9.1995	5400	3.8059	0.9463	0.9998
3.3904	9.3700	5500	3.5559	0.9681	1.0
3.3925	9.5405	5600	3.5007	0.9711	1.0
3.4002	9.7110	5700	3.8154	0.9651	0.9998
3.3646	9.8815	5800	3.7271	0.9651	0.9974
3.3481	10.0512	5900	3.6049	0.9682	1.0
3.3493	10.2217	6000	3.6880	0.9645	1.0
3.3517	10.3922	6100	3.6931	0.9594	0.9995
3.3389	10.5627	6200	3.6221	0.9658	0.9998
3.351	10.7332	6300	3.5192	0.9643	0.9988
3.3732	10.9037	6400	3.8387	0.9562	0.9933
3.3355	11.0733	6500	3.5540	0.9679	1.0
3.3241	11.2438	6600	3.9117	0.9568	1.0
3.325	11.4143	6700	3.4597	0.9684	0.9998
3.3182	11.5848	6800	3.7552	0.9560	0.9944
3.3009	11.7553	6900	3.7510	0.9550	0.9875
3.3061	11.9258	7000	3.7513	0.9502	0.9998
3.2736	12.0955	7100	4.0637	0.9594	0.9972
3.3196	12.2660	7200	4.0105	0.9431	0.9991
3.2558	12.4365	7300	3.8223	0.9509	0.9998
3.2786	12.6070	7400	3.9672	0.9482	0.9993
3.3058	12.7775	7500	3.9256	0.9558	0.9993
3.2352	12.9480	7600	3.8248	0.9556	0.9849
3.2395	13.1176	7700	3.7745	0.9544	0.9884
3.2468	13.2882	7800	3.7510	0.9536	0.9882
3.2628	13.4587	7900	3.5977	0.9438	0.9868
3.2254	13.6292	8000	3.7842	0.9462	1.0009
3.26	13.7997	8100	3.6298	0.9616	0.9949
3.2314	13.9702	8200	3.5676	0.9558	0.9865
3.2138	14.1398	8300	3.6229	0.9651	0.9998
3.2246	14.3103	8400	4.0188	0.9614	0.9972
3.206	14.4808	8500	4.0960	0.9445	0.9886
3.1805	14.6513	8600	3.5612	0.9658	0.9981
3.2015	14.8218	8700	3.5860	0.9499	0.9974
3.2258	14.9923	8800	3.8274	0.9608	0.9986
3.174	15.1620	8900	3.7214	0.9559	0.9963
3.1546	15.3325	9000	4.3387	0.9361	1.0005
3.1583	15.5030	9100	4.3383	0.9453	0.9974
3.1766	15.6735	9200	4.0330	0.9290	0.9916
3.1202	15.8440	9300	3.9352	0.9368	0.9968
3.1504	16.0136	9400	4.3483	0.9471	0.9972
3.1646	16.1841	9500	4.2858	0.9494	0.9970
3.1224	16.3546	9600	3.7921	0.9543	0.9995
3.112	16.5251	9700	4.2156	0.9512	0.9986
3.1261	16.6957	9800	4.2245	0.9476	0.9965
3.0862	16.8662	9900	4.4306	0.9466	0.9970
3.1029	17.0358	10000	4.3931	0.9485	0.9956
3.1263	17.2063	10100	3.4666	0.9473	0.9944
3.075	17.3768	10200	3.6469	0.9464	0.9896
3.0742	17.5473	10300	3.8756	0.9450	0.9863
3.0918	17.7178	10400	3.8311	0.9480	0.9942
3.0716	17.8883	10500	3.5630	0.9487	0.9972
3.067	18.0580	10600	3.8665	0.9409	0.9879
3.0634	18.2285	10700	3.7174	0.9440	0.9942
3.0698	18.3990	10800	3.6759	0.9504	1.0
3.0336	18.5695	10900	4.0177	0.9330	0.9979
3.0657	18.7400	11000	4.3919	0.9405	0.9877
3.0139	18.9105	11100	3.5794	0.9500	0.9956
3.0357	19.0801	11200	3.8221	0.9331	0.9963
3.0356	19.2506	11300	3.7374	0.9423	0.9863
3.0324	19.4211	11400	3.6663	0.9516	0.9965
3.0064	19.5916	11500	4.1142	0.9504	0.9988
2.9911	19.7621	11600	3.7890	0.9558	0.9995
2.9757	19.9327	11700	4.1830	0.9411	0.9961
2.9811	20.1023	11800	4.2312	0.9315	0.9796
2.969	20.2728	11900	4.1445	0.9375	0.9905
3.0039	20.4433	12000	3.9386	0.9345	0.9807
2.9678	20.6138	12100	4.1448	0.9447	0.9937
2.9596	20.7843	12200	4.4927	0.9388	0.9956
2.9558	20.9548	12300	3.8670	0.9408	0.9884
2.9371	21.1245	12400	3.7713	0.9471	0.9954
2.9397	21.2950	12500	4.0556	0.9461	0.9914
2.9382	21.4655	12600	4.0336	0.9418	0.9842
2.9424	21.6360	12700	4.3215	0.9423	0.9833
2.9207	21.8065	12800	4.4437	0.9348	0.9800
2.9332	21.9770	12900	3.6444	0.9352	0.9872
2.9054	22.1466	13000	3.6124	0.9495	0.9963
2.8989	22.3171	13100	4.1020	0.9452	0.9845
2.9295	22.4876	13200	3.8291	0.9469	0.9921
2.8928	22.6581	13300	3.9756	0.9357	0.9819
2.9169	22.8286	13400	4.5840	0.9305	0.9998
2.876	22.9991	13500	4.3819	0.9394	0.9979
2.8265	23.1688	13600	3.9561	0.9325	0.9896
2.8623	23.3393	13700	4.1166	0.9376	0.9988
2.8554	23.5098	13800	4.4085	0.9337	0.9954
2.8668	23.6803	13900	4.1019	0.9307	0.9986
2.9283	23.8508	14000	3.6885	0.9426	0.9991
2.8797	24.0205	14100	4.2674	0.9378	1.0
2.861	24.1910	14200	4.0909	0.9388	0.9991
2.8567	24.3615	14300	4.6243	0.9376	0.9986
2.8289	24.5320	14400	4.4921	0.9323	0.9998
2.8416	24.7025	14500	4.2127	0.9405	0.9968
2.8427	24.8730	14600	3.6274	0.9422	0.9974
2.8449	25.0426	14700	3.7353	0.9424	0.9991
2.8254	25.2131	14800	4.1203	0.9357	0.9986
2.8269	25.3836	14900	4.2943	0.9405	0.9979
2.7912	25.5541	15000	4.2259	0.9344	0.9928
2.8326	25.7246	15100	4.5693	0.9380	0.9933
2.7957	25.8951	15200	4.6771	0.9349	0.9965
2.813	26.0648	15300	3.8681	0.9422	0.9993
2.7835	26.2353	15400	3.9947	0.9392	0.9998
2.8396	26.4058	15500	4.2958	0.9381	0.9984
2.787	26.5763	15600	4.6321	0.9280	0.9991
2.7925	26.7468	15700	3.9910	0.9375	0.9993
2.7928	26.9173	15800	4.0405	0.9379	0.9977
2.7983	27.0870	15900	4.2174	0.9355	0.9940
2.7935	27.2575	16000	4.0693	0.9364	0.9944
2.7497	27.4280	16100	4.0925	0.9370	0.9961
2.7529	27.5985	16200	4.1622	0.9353	0.9974
2.753	27.7690	16300	4.0096	0.9395	0.9974
2.8012	27.9395	16400	4.2769	0.9353	0.9968
2.7372	28.1091	16500	4.3867	0.9359	0.9993
2.749	28.2796	16600	4.1940	0.9388	0.9963
2.741	28.4501	16700	4.1484	0.9390	0.9984
2.7541	28.6206	16800	4.0567	0.9374	0.9988
2.7654	28.7911	16900	4.0987	0.9393	0.9974
2.7408	28.9616	17000	3.9829	0.9402	0.9991
2.7754	29.1313	17100	4.1334	0.9410	0.9977
2.7147	29.3018	17200	4.1019	0.9401	0.9991
2.739	29.4723	17300	4.1809	0.9395	0.9986
2.7449	29.6428	17400	4.1195	0.9385	0.9979
2.7468	29.8133	17500	4.0709	0.9377	0.9965
2.7354	29.9838	17600	4.0536	0.9382	0.9965

Framework versions

Transformers 4.57.2
Pytorch 2.9.1+cu128
Datasets 3.6.0
Tokenizers 0.22.0

ctaguchi/ssc-mmc-model

ssc-mmc-model

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

ctaguchi/
ssc-mmc-model