The model is trained considering the annotation of annotators from Germany only, on instances in German (all linguistic varieties). The annotations from these annotators are aggregated using majority voting and then used to train the model.

Training and evaluation data

The model has been trained on the annotation from annotators from Germany from the MultiPICo dataset. The data has been randomly split into a train and a validation set.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-06
train_batch_size: 16
eval_batch_size: 64
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.0044	1.0	85	0.0043	0.6372	0.3533	0.4417	0.3926
0.0041	2.0	170	0.0041	0.6571	0.3856	0.4917	0.4322
0.0039	3.0	255	0.0039	0.5619	0.3274	0.6167	0.4277
0.0035	4.0	340	0.0037	0.5951	0.3478	0.6	0.4404
0.0033	5.0	425	0.0037	0.5885	0.3429	0.6	0.4364
0.0026	6.0	510	0.0037	0.6128	0.3659	0.625	0.4615
0.0025	7.0	595	0.0035	0.5177	0.3367	0.8417	0.4810
0.0021	8.0	680	0.0036	0.5664	0.3504	0.7417	0.4759
0.0013	9.0	765	0.0040	0.5332	0.3392	0.8	0.4764

Framework versions

Transformers 4.34.1
Pytorch 2.0.1+cu117
Datasets 2.14.5
Tokenizers 0.14.1

Multilingual-Perspectivist-NLU/irony_de_Germany

irony_de_Germany

Model description

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Multilingual-Perspectivist-NLU/
irony_de_Germany