muk-english-digits-classification

This model is a fine-tuned version of facebook/w2v-bert-2.0 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8012
• Accuracy: 1.0

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 1
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.3974	1.0	32	2.3805	0.1875
2.3542	2.0	64	2.3080	0.1875
2.2986	3.0	96	2.1997	0.3438
2.1947	4.0	128	2.0517	0.4062
2.0741	5.0	160	1.9182	0.4062
1.9479	6.0	192	1.8475	0.4688
1.8934	7.0	224	1.6724	0.5938
1.687	8.0	256	1.5422	0.6875
1.5745	9.0	288	1.3878	0.875
1.5288	10.0	320	1.2905	0.875
1.3518	11.0	352	1.1875	0.9375
1.2558	12.0	384	1.0936	0.9375
1.1538	13.0	416	1.0296	0.9688
1.186	14.0	448	0.9815	0.9688
1.033	15.0	480	0.9271	1.0
0.9828	16.0	512	0.8811	1.0
0.995	17.0	544	0.8454	0.9688
0.9213	18.0	576	0.8176	1.0
0.9313	19.0	608	0.8048	1.0
0.9029	20.0	640	0.8012	1.0

Framework versions

• Transformers 4.57.2
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1