output_model_shunyalabs_data_base_model_more_steps

This model is a fine-tuned version of Eimhin03/output_model_shunyalabs_data_base_model on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8582
• Wer: 28.5187

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: 0.0001
• train_batch_size: 2
• eval_batch_size: 4
• seed: 42
• 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: 500
• training_steps: 100000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0443	7.8125	10000	0.8899	40.4519
0.0167	15.625	20000	0.9089	36.5973
0.0206	23.4375	30000	0.9090	34.8545
0.0031	31.25	40000	0.8985	34.9284
0.0016	39.0625	50000	0.9108	34.0718
0.0008	46.875	60000	0.8898	32.7426
0.0004	54.6875	70000	0.8897	30.7931
0.0003	62.5	80000	0.8849	30.2023
0.0000	70.3125	90000	0.8603	29.9365
0.0000	78.125	100000	0.8582	28.5187

Framework versions

• Transformers 5.2.0.dev0
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.2