You can deploy this model and start earning money today!

nik1509/
xlsr300m_hiddendropout_ablation_hd_03

Name: xlsr300m_hiddendropout_ablation_hd_03
Availability: InStock

Views

No views yet

xlsr300m_hiddendropout_ablation_hd_03

This model is a fine-tuned version of nik1509/xlsr300m_finetuned_telugu_basemodel on the None dataset. It achieves the following results on the evaluation set:

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Use 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_ratio: 0.1
num_epochs: 15
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
336.9262	0.6076	700	452.3805	0.4025	0.1672
285.0287	1.2153	1400	459.2792	0.4006	0.1663
300.8956	1.8229	2100	447.6531	0.4079	0.1687
297.0937	2.4306	2800	461.8190	0.4044	0.1676
305.9566	3.0382	3500	462.1948	0.4020	0.1674
241.0523	3.6458	4200	472.9764	0.4030	0.1679
267.8165	4.2535	4900	474.1494	0.4050	0.1686
305.6236	4.8611	5600	476.0005	0.4047	0.1675
270.1654	5.4688	6300	483.4022	0.4050	0.1685
280.894	6.0764	7000	493.5164	0.4004	0.1677
287.6753	6.6840	7700	491.2645	0.3980	0.1667
295.4928	7.2917	8400	496.6277	0.4006	0.1667
285.4525	7.8993	9100	493.0818	0.3991	0.1674
238.2578	8.5069	9800	509.1539	0.3982	0.1672
257.4116	9.1146	10500	513.8447	0.3993	0.1672
303.4367	9.7222	11200	510.1010	0.3978	0.1670
267.1826	10.3299	11900	524.4132	0.3958	0.1669
249.4068	10.9375	12600	518.0434	0.3947	0.1666
281.035	11.5451	13300	522.9650	0.3995	0.1669
234.234	12.1528	14000	529.6373	0.3969	0.1666
286.6428	12.7604	14700	531.5583	0.3975	0.1667
250.7785	13.3681	15400	534.5762	0.3955	0.1666
280.5476	13.9757	16100	530.7258	0.3959	0.1665
229.8652	14.5833	16800	532.1289	0.3968	0.1667