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

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

• Loss: 0.2936
• Precision: 0.6330
• Recall: 0.8133
• F1: 0.7119
• Accuracy: 0.9646

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	0.2534	0.1244	0.4813	0.1977	0.8342
No log	2.0	344	0.1838	0.1965	0.5274	0.2863	0.8817
0.4271	3.0	516	0.1549	0.2364	0.5400	0.3289	0.9107
0.4271	4.0	688	0.2007	0.2979	0.5944	0.3969	0.9146
0.4271	5.0	860	0.1944	0.3316	0.6133	0.4304	0.9252
0.1111	6.0	1032	0.2037	0.3317	0.6213	0.4325	0.9226
0.1111	7.0	1204	0.2115	0.3849	0.6695	0.4888	0.9379
0.1111	8.0	1376	0.2312	0.3921	0.6897	0.5	0.9353
0.0595	9.0	1548	0.2203	0.3655	0.6248	0.4612	0.9362
0.0595	10.0	1720	0.2458	0.5016	0.7550	0.6028	0.9534
0.0595	11.0	1892	0.2598	0.4284	0.6866	0.5276	0.9464
0.0361	12.0	2064	0.2471	0.5127	0.7445	0.6073	0.9550
0.0361	13.0	2236	0.2719	0.5211	0.7407	0.6118	0.9561
0.0361	14.0	2408	0.2611	0.5459	0.7640	0.6368	0.9571
0.0216	15.0	2580	0.2675	0.5965	0.8126	0.6879	0.9615
0.0216	16.0	2752	0.2775	0.5994	0.8073	0.6880	0.9611
0.0216	17.0	2924	0.2776	0.6226	0.8080	0.7033	0.9636
0.0142	18.0	3096	0.2876	0.6299	0.8115	0.7093	0.9646
0.0142	19.0	3268	0.2885	0.6373	0.8150	0.7153	0.9648
0.0142	20.0	3440	0.2936	0.6330	0.8133	0.7119	0.9646

Framework versions

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