facebook-roberta-large-finetuned-ner-vlsp2021-3090-14June

This model is a fine-tuned version of FacebookAI/xlm-roberta-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0999
• Atetime: {'precision': 0.8815399802566634, 'recall': 0.8912175648702595, 'f1': 0.8863523573200993, 'number': 1002}
• Ddress: {'precision': 0.7941176470588235, 'recall': 0.9310344827586207, 'f1': 0.8571428571428571, 'number': 29}
• Erson: {'precision': 0.9600840336134454, 'recall': 0.9626119010005266, 'f1': 0.9613463055482514, 'number': 1899}
• Ersontype: {'precision': 0.7473684210526316, 'recall': 0.7266081871345029, 'f1': 0.736842105263158, 'number': 684}
• Honenumber: {'precision': 0.8888888888888888, 'recall': 0.8888888888888888, 'f1': 0.8888888888888888, 'number': 9}
• Iscellaneous: {'precision': 0.5126582278481012, 'recall': 0.5094339622641509, 'f1': 0.5110410094637223, 'number': 159}
• Mail: {'precision': 1.0, 'recall': 0.9803921568627451, 'f1': 0.99009900990099, 'number': 51}
• Ocation: {'precision': 0.8736842105263158, 'recall': 0.8931591083781706, 'f1': 0.8833143291524135, 'number': 1301}
• P: {'precision': 0.75, 'recall': 0.8181818181818182, 'f1': 0.7826086956521738, 'number': 11}
• Rl: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 15}
• Roduct: {'precision': 0.7059773828756059, 'recall': 0.6992, 'f1': 0.702572347266881, 'number': 625}
• Overall Precision: 0.8619
• Overall Recall: 0.8655
• Overall F1: 0.8637
• Overall Accuracy: 0.9810

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Atetime	Ddress	Erson	Ersontype	Honenumber	Iscellaneous	Mail	Ocation	P	Rl	Roduct	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.097	1.0	3263	0.0814	{'precision': 0.8521825396825397, 'recall': 0.8572854291417166, 'f1': 0.854726368159204, 'number': 1002}	{'precision': 0.574468085106383, 'recall': 0.9310344827586207, 'f1': 0.7105263157894737, 'number': 29}	{'precision': 0.9606681034482759, 'recall': 0.9389152185360716, 'f1': 0.9496671105193077, 'number': 1899}	{'precision': 0.7643097643097643, 'recall': 0.6637426900584795, 'f1': 0.7104851330203442, 'number': 684}	{'precision': 0.7272727272727273, 'recall': 0.8888888888888888, 'f1': 0.7999999999999999, 'number': 9}	{'precision': 0.4517766497461929, 'recall': 0.559748427672956, 'f1': 0.5, 'number': 159}	{'precision': 1.0, 'recall': 0.9411764705882353, 'f1': 0.9696969696969697, 'number': 51}	{'precision': 0.8556390977443609, 'recall': 0.8747117601844735, 'f1': 0.8650703154694033, 'number': 1301}	{'precision': 0.6666666666666666, 'recall': 0.7272727272727273, 'f1': 0.6956521739130435, 'number': 11}	{'precision': 0.7222222222222222, 'recall': 0.8666666666666667, 'f1': 0.7878787878787877, 'number': 15}	{'precision': 0.5155555555555555, 'recall': 0.5568, 'f1': 0.5353846153846155, 'number': 625}	0.8238	0.8254	0.8246	0.9769
0.0596	2.0	6526	0.0905	{'precision': 0.8613569321533924, 'recall': 0.874251497005988, 'f1': 0.8677563150074294, 'number': 1002}	{'precision': 0.5111111111111111, 'recall': 0.7931034482758621, 'f1': 0.6216216216216216, 'number': 29}	{'precision': 0.967741935483871, 'recall': 0.9478672985781991, 'f1': 0.9577015163607343, 'number': 1899}	{'precision': 0.8167770419426048, 'recall': 0.5409356725146199, 'f1': 0.6508355321020229, 'number': 684}	{'precision': 0.8, 'recall': 0.8888888888888888, 'f1': 0.8421052631578948, 'number': 9}	{'precision': 0.51875, 'recall': 0.5220125786163522, 'f1': 0.5203761755485894, 'number': 159}	{'precision': 1.0, 'recall': 0.9607843137254902, 'f1': 0.98, 'number': 51}	{'precision': 0.8402323892519971, 'recall': 0.889315910837817, 'f1': 0.8640776699029126, 'number': 1301}	{'precision': 0.5833333333333334, 'recall': 0.6363636363636364, 'f1': 0.6086956521739131, 'number': 11}	{'precision': 0.5555555555555556, 'recall': 0.6666666666666666, 'f1': 0.606060606060606, 'number': 15}	{'precision': 0.6890595009596929, 'recall': 0.5744, 'f1': 0.6265270506108203, 'number': 625}	0.8587	0.8197	0.8388	0.9779
0.0395	3.0	9789	0.0885	{'precision': 0.8682877406281662, 'recall': 0.8552894211576846, 'f1': 0.8617395676219206, 'number': 1002}	{'precision': 0.6, 'recall': 0.8275862068965517, 'f1': 0.6956521739130435, 'number': 29}	{'precision': 0.9590643274853801, 'recall': 0.9499736703528173, 'f1': 0.9544973544973545, 'number': 1899}	{'precision': 0.7365930599369085, 'recall': 0.6827485380116959, 'f1': 0.7086494688922609, 'number': 684}	{'precision': 0.8, 'recall': 0.8888888888888888, 'f1': 0.8421052631578948, 'number': 9}	{'precision': 0.5212121212121212, 'recall': 0.5408805031446541, 'f1': 0.5308641975308641, 'number': 159}	{'precision': 1.0, 'recall': 0.9607843137254902, 'f1': 0.98, 'number': 51}	{'precision': 0.8641881638846738, 'recall': 0.8754803996925442, 'f1': 0.8697976326842306, 'number': 1301}	{'precision': 0.75, 'recall': 0.8181818181818182, 'f1': 0.7826086956521738, 'number': 11}	{'precision': 0.7647058823529411, 'recall': 0.8666666666666667, 'f1': 0.8125, 'number': 15}	{'precision': 0.6871880199667221, 'recall': 0.6608, 'f1': 0.6737357259380099, 'number': 625}	0.8521	0.8417	0.8469	0.9794
0.0248	4.0	13052	0.0953	{'precision': 0.8802395209580839, 'recall': 0.8802395209580839, 'f1': 0.8802395209580839, 'number': 1002}	{'precision': 0.7428571428571429, 'recall': 0.896551724137931, 'f1': 0.8125, 'number': 29}	{'precision': 0.9623741388447271, 'recall': 0.956292785676672, 'f1': 0.9593238246170102, 'number': 1899}	{'precision': 0.7728758169934641, 'recall': 0.6915204678362573, 'f1': 0.7299382716049383, 'number': 684}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 9}	{'precision': 0.4857142857142857, 'recall': 0.5345911949685535, 'f1': 0.5089820359281437, 'number': 159}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 51}	{'precision': 0.8614591009579956, 'recall': 0.8985395849346657, 'f1': 0.8796087283671934, 'number': 1301}	{'precision': 0.8181818181818182, 'recall': 0.8181818181818182, 'f1': 0.8181818181818182, 'number': 11}	{'precision': 0.9333333333333333, 'recall': 0.9333333333333333, 'f1': 0.9333333333333333, 'number': 15}	{'precision': 0.7482876712328768, 'recall': 0.6992, 'f1': 0.7229114971050455, 'number': 625}	0.8663	0.8593	0.8628	0.9806
0.0161	5.0	16315	0.0999	{'precision': 0.8815399802566634, 'recall': 0.8912175648702595, 'f1': 0.8863523573200993, 'number': 1002}	{'precision': 0.7941176470588235, 'recall': 0.9310344827586207, 'f1': 0.8571428571428571, 'number': 29}	{'precision': 0.9600840336134454, 'recall': 0.9626119010005266, 'f1': 0.9613463055482514, 'number': 1899}	{'precision': 0.7473684210526316, 'recall': 0.7266081871345029, 'f1': 0.736842105263158, 'number': 684}	{'precision': 0.8888888888888888, 'recall': 0.8888888888888888, 'f1': 0.8888888888888888, 'number': 9}	{'precision': 0.5126582278481012, 'recall': 0.5094339622641509, 'f1': 0.5110410094637223, 'number': 159}	{'precision': 1.0, 'recall': 0.9803921568627451, 'f1': 0.99009900990099, 'number': 51}	{'precision': 0.8736842105263158, 'recall': 0.8931591083781706, 'f1': 0.8833143291524135, 'number': 1301}	{'precision': 0.75, 'recall': 0.8181818181818182, 'f1': 0.7826086956521738, 'number': 11}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 15}	{'precision': 0.7059773828756059, 'recall': 0.6992, 'f1': 0.702572347266881, 'number': 625}	0.8619	0.8655	0.8637	0.9810

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1