Modelpublic

nguyenkhoa2407/roberta-base-NER-favsbot

sourceHugging Facemitupdated 4y agoView on Hugging Face

0likes14downloads

Model Card

roberta-base-NER-favsbot

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

—Loss: 0.4812
—Precision: 0.6940
—Recall: 0.7056
—F1: 0.6997
—Accuracy: 0.8454

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: 1.5e-05
—trainbatchsize: 16
—evalbatchsize: 16
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	4	2.1885	0.1787	0.3722	0.2414	0.2998
No log	2.0	8	1.9109	0.3768	0.1444	0.2088	0.4707
No log	3.0	12	1.6425	0.0	0.0	0.0	0.4075
No log	4.0	16	1.5402	0.75	0.0167	0.0326	0.4145
No log	5.0	20	1.3569	0.5263	0.3333	0.4082	0.5785
No log	6.0	24	1.2250	0.4691	0.5056	0.4866	0.6721
No log	7.0	28	1.0957	0.4973	0.5167	0.5068	0.6885
No log	8.0	32	0.9751	0.5581	0.5333	0.5455	0.7190
No log	9.0	36	0.8824	0.6456	0.5667	0.6036	0.7518
No log	10.0	40	0.8035	0.6442	0.5833	0.6122	0.7658
No log	11.0	44	0.7336	0.6412	0.6056	0.6229	0.7822
No log	12.0	48	0.6722	0.6384	0.6278	0.6331	0.7963
No log	13.0	52	0.6205	0.6610	0.65	0.6555	0.8080
No log	14.0	56	0.5796	0.6740	0.6778	0.6759	0.8244
No log	15.0	60	0.5503	0.6796	0.6833	0.6814	0.8244
No log	16.0	64	0.5256	0.6851	0.6889	0.6870	0.8337
No log	17.0	68	0.5066	0.6868	0.6944	0.6906	0.8407
No log	18.0	72	0.4929	0.6868	0.6944	0.6906	0.8431
No log	19.0	76	0.4845	0.6868	0.6944	0.6906	0.8431
No log	20.0	80	0.4812	0.6940	0.7056	0.6997	0.8454

Framework versions

—Transformers 4.21.1
—Pytorch 1.12.1
—Datasets 2.4.0
—Tokenizers 0.12.1