Modelpublic

dmargutierrez/distilbert-base-uncased-mapa-ner-coarse_grained-v2

sourceHugging Faceapache-2.0updated 4y agoView on Hugging Face

0likes9downloads

Model Card

distilbert-base-uncased-mapa-ner-coarse_grained-v2

This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.1627
—Precision: 0.7898
—Recall: 0.4843
—F1: 0.6004
—Accuracy: 0.9857

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
—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
0.0354	1.0	1739	0.0834	0.7023	0.4612	0.5568	0.9840
0.0255	2.0	3478	0.1034	0.8172	0.4355	0.5682	0.9852
0.0168	3.0	5217	0.0969	0.7714	0.4588	0.5754	0.9848
0.0132	4.0	6956	0.1042	0.7477	0.4838	0.5875	0.9852
0.0112	5.0	8695	0.1109	0.7421	0.4863	0.5876	0.9849
0.0085	6.0	10434	0.1076	0.7194	0.4951	0.5865	0.9850
0.0067	7.0	12173	0.1343	0.7828	0.4587	0.5784	0.9849
0.0047	8.0	13912	0.1252	0.7425	0.4840	0.5860	0.9853
0.0045	9.0	15651	0.1410	0.7943	0.4615	0.5838	0.9852
0.0035	10.0	17390	0.1311	0.7624	0.4929	0.5987	0.9857
0.003	11.0	19129	0.1494	0.8059	0.4691	0.5930	0.9855
0.0025	12.0	20868	0.1436	0.7674	0.4852	0.5945	0.9856
0.002	13.0	22607	0.1513	0.7778	0.4741	0.5891	0.9852
0.0014	14.0	24346	0.1577	0.7986	0.4726	0.5938	0.9855
0.0016	15.0	26085	0.1573	0.7802	0.4766	0.5917	0.9855
0.0011	16.0	27824	0.1599	0.7917	0.4723	0.5916	0.9856
0.0012	17.0	29563	0.1601	0.7848	0.4867	0.6008	0.9857
0.001	18.0	31302	0.1572	0.7614	0.4939	0.5991	0.9856
0.0011	19.0	33041	0.1602	0.7858	0.4870	0.6013	0.9857
0.0009	20.0	34780	0.1627	0.7898	0.4843	0.6004	0.9857

Framework versions

—Transformers 4.26.0
—Pytorch 1.13.1+cu117
—Datasets 2.9.0
—Tokenizers 0.13.2