metadata

base_model: FacebookAI/xlm-roberta-base
library_name: transformers
license: mit
metrics:
  - precision
  - recall
  - f1
  - accuracy
tags:
  - generated_from_trainer
model-index:
  - name: scenario-non-kd-pre-ner-full-xlmr_data-univner_en44
    results: []

scenario-non-kd-pre-ner-full-xlmr_data-univner_en44

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

Loss: 0.1507
Precision: 0.7345
Recall: 0.7588
F1: 0.7464
Accuracy: 0.9803

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: 32
eval_batch_size: 32
seed: 44
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0869	1.2755	500	0.0715	0.6808	0.7308	0.7049	0.9773
0.0328	2.5510	1000	0.0781	0.7035	0.7319	0.7174	0.9784
0.0202	3.8265	1500	0.0805	0.7083	0.7340	0.7209	0.9791
0.0121	5.1020	2000	0.0914	0.7255	0.7495	0.7373	0.9792
0.0071	6.3776	2500	0.0988	0.7178	0.7557	0.7363	0.9787
0.005	7.6531	3000	0.1089	0.7178	0.7609	0.7387	0.9795
0.0033	8.9286	3500	0.1177	0.7353	0.7246	0.7299	0.9792
0.0033	10.2041	4000	0.1134	0.7219	0.7391	0.7304	0.9794
0.0022	11.4796	4500	0.1251	0.7243	0.7588	0.7412	0.9801
0.0023	12.7551	5000	0.1279	0.7070	0.7619	0.7334	0.9792
0.0017	14.0306	5500	0.1231	0.7165	0.7588	0.7371	0.9793
0.0014	15.3061	6000	0.1378	0.7289	0.7598	0.7440	0.9792
0.0014	16.5816	6500	0.1507	0.6986	0.7774	0.7359	0.9782
0.001	17.8571	7000	0.1421	0.7242	0.7474	0.7356	0.9793
0.0009	19.1327	7500	0.1396	0.7284	0.7578	0.7428	0.9796
0.0008	20.4082	8000	0.1394	0.7402	0.7226	0.7313	0.9788
0.0006	21.6837	8500	0.1425	0.7542	0.7371	0.7455	0.9797
0.0006	22.9592	9000	0.1446	0.7339	0.7308	0.7324	0.9793
0.0005	24.2347	9500	0.1459	0.7374	0.7557	0.7464	0.9804
0.0004	25.5102	10000	0.1443	0.7323	0.7505	0.7413	0.9800
0.0002	26.7857	10500	0.1485	0.7299	0.7526	0.7411	0.9801
0.0003	28.0612	11000	0.1507	0.7408	0.7484	0.7446	0.9803
0.0003	29.3367	11500	0.1507	0.7345	0.7588	0.7464	0.9803

Framework versions

Transformers 4.44.2
Pytorch 2.1.1+cu121
Datasets 2.14.5
Tokenizers 0.19.1