bert-finetuned-goodsmemo-ner

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

• Loss: 0.1899
• Precision: 0.1455
• Recall: 0.1495
• F1: 0.1475
• Accuracy: 0.9294

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: 24
• eval_batch_size: 24
• 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	46	0.3317	0.0	0.0	0.0	0.9018
No log	2.0	92	0.3051	0.0090	0.0280	0.0137	0.8640
No log	3.0	138	0.2561	0.0207	0.0467	0.0287	0.8966
No log	4.0	184	0.2345	0.0383	0.0748	0.0506	0.9118
No log	5.0	230	0.2319	0.0491	0.1028	0.0665	0.9018
No log	6.0	276	0.2108	0.1085	0.1308	0.1186	0.9245
No log	7.0	322	0.2042	0.1181	0.1402	0.1282	0.9268
No log	8.0	368	0.2077	0.1262	0.1215	0.1238	0.9263
No log	9.0	414	0.1951	0.1524	0.1495	0.1509	0.9297
No log	10.0	460	0.1899	0.1455	0.1495	0.1475	0.9294

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu117
• Datasets 2.14.4
• Tokenizers 0.13.3