bert-large-cased-upos

This model is a fine-tuned version of bert-large-cased on the universal_dependencies dataset. It achieves the following results on the evaluation set:

• Loss: 0.4114
• Precision: 0.8688
• Recall: 0.8557
• F1: 0.8618
• Accuracy: 0.8904

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	438	0.5774	0.8348	0.7595	0.7893	0.8099
No log	2.0	876	0.4787	0.8114	0.7967	0.8000	0.8385
No log	3.0	1314	0.4345	0.8227	0.8302	0.8213	0.8601
No log	4.0	1752	0.4140	0.8257	0.8430	0.8304	0.8727
No log	5.0	2190	0.4211	0.8405	0.8525	0.8441	0.8787
No log	6.0	2628	0.4114	0.8688	0.8557	0.8618	0.8904
No log	7.0	3066	0.4582	0.8454	0.8572	0.8503	0.8911
No log	8.0	3504	0.4771	0.8447	0.8588	0.8508	0.8894
No log	9.0	3942	0.4799	0.8545	0.8626	0.8577	0.8918
No log	10.0	4380	0.4919	0.8539	0.8642	0.8579	0.8937

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1