bert-question-ner

This model is a fine-tuned version of google-bert/bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3300
• Precision: 0.7339
• Recall: 0.7477
• F1: 0.7407
• Accuracy: 0.9154

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• 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_ratio: 0.4
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	68	1.5591	0.0203	0.0187	0.0195	0.6473
No log	2.0	136	0.8601	0.8333	0.0467	0.0885	0.7313
No log	3.0	204	0.5763	0.4365	0.3692	0.4	0.8133
No log	4.0	272	0.4647	0.5294	0.5888	0.5575	0.8650
No log	5.0	340	0.3700	0.5935	0.6822	0.6348	0.8960
No log	6.0	408	0.3458	0.6872	0.7290	0.7075	0.9050
No log	7.0	476	0.3007	0.7207	0.7477	0.7339	0.9134
0.7147	8.0	544	0.3036	0.7302	0.7336	0.7319	0.9109
0.7147	9.0	612	0.2997	0.7260	0.7430	0.7344	0.9160
0.7147	10.0	680	0.3247	0.6964	0.7290	0.7123	0.9141
0.7147	11.0	748	0.3300	0.7339	0.7477	0.7407	0.9154
0.7147	12.0	816	0.3582	0.7162	0.7430	0.7294	0.9154
0.7147	13.0	884	0.3702	0.7281	0.7383	0.7332	0.9154

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0