ssr-360m

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0227
• Model Preparation Time: 0.0066
• Wer: 0.1206

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: 0.0005
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 10
• total_train_batch_size: 20
• 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
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Wer
12.0652	0.9999	5200	0.5135	0.0066	0.4332
2.1537	1.9999	10400	0.0530	0.0066	0.1585
3.0568	2.9999	15600	0.0731	0.0066	0.2030
0.9142	3.9999	20800	0.0313	0.0066	0.1407
0.3436	4.9999	26000	0.0308	0.0066	0.1403
0.7558	5.9999	31200	0.0286	0.0066	0.1337
0.5733	6.9999	36400	0.0265	0.0066	0.1221
0.1573	7.9999	41600	0.0241	0.0066	0.1439
0.2248	8.9999	46800	0.0225	0.0066	0.1199
0.1501	9.9999	52000	0.0227	0.0066	0.1206

Framework versions

• Transformers 4.49.0
• Pytorch 2.5.1
• Datasets 3.3.2
• Tokenizers 0.21.0