xmod-shared-roberta-base-legal-multi-downstream-indian-ner

This model is a fine-tuned version of MHGanainy/xmod-shared-roberta-base-legal-multi on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2694
• Precision: 0.6081
• Recall: 0.8059
• F1: 0.6932
• Accuracy: 0.9644

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: 3e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 1
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	172	0.2460	0.1302	0.4831	0.2052	0.8390
No log	2.0	344	0.1802	0.1852	0.5113	0.2719	0.8825
0.4295	3.0	516	0.1729	0.2255	0.5459	0.3191	0.8990
0.4295	4.0	688	0.1849	0.3068	0.6007	0.4061	0.9191
0.4295	5.0	860	0.1756	0.2794	0.5871	0.3786	0.9135
0.1145	6.0	1032	0.1931	0.3640	0.6698	0.4717	0.9326
0.1145	7.0	1204	0.2178	0.3763	0.6677	0.4813	0.9344
0.1145	8.0	1376	0.2150	0.3793	0.6925	0.4902	0.9326
0.0604	9.0	1548	0.2101	0.3977	0.6600	0.4963	0.9385
0.0604	10.0	1720	0.2132	0.4825	0.7187	0.5774	0.9511
0.0604	11.0	1892	0.2265	0.4761	0.7361	0.5782	0.9494
0.0385	12.0	2064	0.2415	0.5150	0.7483	0.6101	0.9536
0.0385	13.0	2236	0.2358	0.5077	0.7504	0.6056	0.9553
0.0385	14.0	2408	0.2509	0.5203	0.7543	0.6158	0.9546
0.0224	15.0	2580	0.2398	0.5683	0.8028	0.6655	0.9593
0.0224	16.0	2752	0.2665	0.5949	0.8115	0.6865	0.9613
0.0224	17.0	2924	0.2543	0.5874	0.8070	0.6799	0.9617
0.0156	18.0	3096	0.2672	0.5862	0.7951	0.6749	0.9620
0.0156	19.0	3268	0.2658	0.6002	0.8021	0.6866	0.9632
0.0156	20.0	3440	0.2694	0.6081	0.8059	0.6932	0.9644

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1