whisper-small-pashto-run11-epoch2

This model is a fine-tuned version of Adnan666/whisper-small-pashto-run11-cv24only on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.0639
Wer: 50.2334

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-06
train_batch_size: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH 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: 200
num_epochs: 1
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1845	0.0080	500	1.0338	50.7937
0.1866	0.0160	1000	1.0318	49.3744
0.1770	0.0240	1500	1.0183	49.3931
0.1579	0.0320	2000	0.9842	49.1317
0.1726	0.0400	2500	1.0034	48.1232
0.2272	0.0480	3000	1.0191	48.8142
0.1434	0.0560	3500	1.0447	49.2997
0.1544	0.0640	4000	1.0274	49.6545
0.1801	0.0720	4500	1.0476	49.6172
0.1356	0.0800	5000	1.0478	50.2334
0.1549	0.0880	5500	1.0228	50.1214
0.1612	0.0960	6000	1.0229	49.4678
0.1210	0.1040	6500	1.0701	50.4575
0.1522	0.1120	7000	1.0692	50.5696
0.1404	0.1200	7500	1.0634	50.1587
0.1511	0.1280	8000	1.1004	51.3725
0.1131	0.1360	8500	1.0945	51.0924
0.1137	0.1440	9000	1.0897	50.2894
0.1167	0.1520	9500	1.0755	50.2894
0.1174	0.1600	10000	1.0706	50.3828
0.1395	0.1680	10500	1.1012	50.9057
0.0995	0.1760	11000	1.0722	49.3744
0.0995	0.1840	11500	1.1017	49.9720
0.0920	0.1920	12000	1.0705	49.1317
0.0991	0.2000	12500	1.0929	50.2148
0.0967	0.2080	13000	1.0533	49.4865
0.0971	0.2160	13500	1.0681	49.2250
0.1039	0.2240	14000	1.1135	50.4949
0.0949	0.2320	14500	1.0887	49.9160
0.1145	0.2400	15000	1.0695	49.3557
0.0976	0.2480	15500	1.1011	49.2810
0.0879	0.2560	16000	1.1035	50.4015
0.0823	0.2640	16500	1.0848	48.9449
0.0770	0.2720	17000	1.0651	48.9636
0.1408	0.2800	17500	1.0247	48.7395
0.1573	0.2880	18000	1.0124	47.9365
0.1740	0.2960	18500	1.0292	49.0756
0.1393	0.3040	19000	1.0550	49.8039
0.0890	0.3120	19500	1.0722	50.3081
0.1307	0.3200	20000	1.0725	50.3828
0.1499	0.3280	20500	1.0438	49.7106
0.2019	0.3360	21000	1.0669	50.8870
0.2074	0.3440	21500	1.0505	50.2334
0.1095	0.3520	22000	1.0528	49.5051
0.1869	0.3600	22500	1.0591	50.3828
0.1644	0.3680	23000	1.0672	50.3081
0.1174	0.3760	23500	1.0608	48.7021
0.1193	0.3840	24000	1.0600	49.1317
0.1191	0.3920	24500	1.0759	49.3557
0.1641	0.4000	25000	1.0829	51.1671
0.1678	0.4080	25500	1.0730	50.0654
0.1929	0.4160	26000	1.0729	50.2148
0.1218	0.4240	26500	1.0505	49.3371
0.1587	0.4320	27000	1.0698	50.1774
0.1504	0.4400	27500	1.0723	50.6069
0.1593	0.4480	28000	1.0701	50.4575
0.1979	0.4560	28500	1.0573	49.8413
0.1397	0.4640	29000	1.0544	49.9907
0.2017	0.4720	29500	1.0461	49.5425
0.1399	0.4800	30000	1.0747	50.5882
0.2291	0.4880	30500	1.0657	50.1027
0.1023	0.4960	31000	1.0785	49.8226
0.1717	0.5040	31500	1.0795	49.8786
0.2043	0.5120	32000	1.1049	51.2232
0.1535	0.5200	32500	1.0842	50.7750
0.1955	0.5280	33000	1.1005	50.9804
0.1344	0.5360	33500	1.0868	50.4202
0.2221	0.5440	34000	1.1013	51.4659
0.1868	0.5520	34500	1.0843	50.9617
0.1403	0.5600	35000	1.0900	50.7003
0.1597	0.5680	35500	1.1086	50.8310
0.2540	0.5760	36000	1.0954	51.7087
0.2213	0.5840	36500	1.0957	51.0738
0.2066	0.5920	37000	1.0927	51.8207
0.2232	0.6000	37500	1.1023	51.5406
0.2527	0.6080	38000	1.0890	52.4743
0.2195	0.6160	38500	1.0933	51.9328
0.3150	0.6240	39000	1.0884	51.1671
0.2382	0.6320	39500	1.0806	51.3352
0.2781	0.6400	40000	1.0556	50.7937
0.2898	0.6480	40500	1.0677	51.5406
0.2489	0.6560	41000	1.0527	50.9804
0.2538	0.6640	41500	1.0472	51.3539
0.2776	0.6720	42000	1.0488	51.1671
0.3341	0.6800	42500	1.0514	51.2045
0.2494	0.6880	43000	1.0531	50.4762
0.2354	0.6960	43500	1.0477	50.2894
0.2527	0.7040	44000	1.0360	49.4118
0.2135	0.7120	44500	1.0538	50.7563
0.2454	0.7200	45000	1.0504	50.6256
0.2503	0.7280	45500	1.0423	49.9533
0.2650	0.7360	46000	1.0395	50.0467
0.2851	0.7440	46500	1.0388	49.8413
0.2696	0.7520	47000	1.0483	50.3641
0.2750	0.7600	47500	1.0435	50.2708
0.1773	0.7680	48000	1.0602	50.4762
0.2198	0.7760	48500	1.0626	50.7190
0.2892	0.7840	49000	1.0611	50.3081
0.2092	0.7920	49500	1.0693	50.7563
0.3374	0.8000	50000	1.0552	50.3641
0.3071	0.8080	50500	1.0447	50.3081
0.3032	0.8160	51000	1.0396	50.2894
0.2589	0.8240	51500	1.0502	49.8226
0.2477	0.8320	52000	1.0573	50.4202
0.2989	0.8400	52500	1.0596	49.9160
0.2460	0.8480	53000	1.0588	50.2148
0.1791	0.8560	53500	1.0584	50.4575
0.3227	0.8640	54000	1.0605	50.3268
0.2212	0.8720	54500	1.0596	50.3828
0.2266	0.8800	55000	1.0523	50.4575
0.2220	0.8880	55500	1.0543	50.1401
0.2061	0.8960	56000	1.0537	49.8413
0.2136	0.9040	56500	1.0599	50.1587
0.1832	0.9120	57000	1.0554	50.0280
0.2858	0.9200	57500	1.0554	50.1027
0.3202	0.9280	58000	1.0610	50.4575
0.2234	0.9360	58500	1.0608	51.0551
0.2643	0.9440	59000	1.0617	50.5696
0.2868	0.9520	59500	1.0623	50.4015
0.2347	0.9600	60000	1.0614	50.4949
0.3083	0.9680	60500	1.0611	50.3641
0.3428	0.9760	61000	1.0630	50.4015
0.2567	0.9840	61500	1.0628	50.5322
0.2505	0.9920	62000	1.0639	50.2334

Framework versions

Transformers 5.0.0
Pytorch 2.6.0+cu124
Datasets 4.8.5
Tokenizers 0.22.2

Adnan666/whisper-small-pashto-run11-epoch2

whisper-small-pashto-run11-epoch2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions