mms-telugu-binary

This model is a fine-tuned version of facebook/mms-300m on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6300
• Accuracy: 0.7455
• F1: 0.7742

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• 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_ratio: 0.1
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.6804	1.0	14	0.6916	0.6364	0.7778
0.6736	2.0	28	0.6765	0.6364	0.7778
0.6407	3.0	42	0.6302	0.7636	0.8060
0.578	4.0	56	0.6086	0.6182	0.6182
0.5428	5.0	70	0.5948	0.6364	0.6429
0.4916	6.0	84	0.6348	0.6	0.5926
0.4702	7.0	98	0.6147	0.6364	0.6429
0.4867	8.0	112	0.6111	0.6364	0.6429
0.4372	9.0	126	0.4891	0.8	0.8358
0.3984	10.0	140	0.5122	0.7455	0.7742
0.4102	11.0	154	0.5403	0.7091	0.7419
0.336	12.0	168	0.6108	0.6909	0.7119
0.3143	13.0	182	0.4587	0.8364	0.8696
0.323	14.0	196	0.6597	0.6909	0.7119
0.2945	15.0	210	0.6919	0.6909	0.7119
0.2852	16.0	224	0.6189	0.7455	0.7742
0.2899	17.0	238	0.6230	0.7273	0.7619
0.2582	18.0	252	0.6303	0.7455	0.7742
0.2653	19.0	266	0.6444	0.7273	0.7541
0.243	20.0	280	0.6300	0.7455	0.7742

Framework versions

• Transformers 4.53.3
• Pytorch 2.9.1+cu128
• Datasets 4.4.1
• Tokenizers 0.21.2