rlt

This model is a fine-tuned version of svalabs/gbert-large-zeroshot-nli on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0381
• F1: 0.9901

Label-Übersetzungen

• 0 Abluft
• 1 Abluftfilter
• 2 Abluftventilator
• 3 Andere
• 4 Aussenluftklappe
• 5 Befeuchter
• 6 Brandschutz
• 7 Brandschutzklappe
• 8 Erhitzer
• 9 Kühler
• 10 Nacherhitzer
• 11 RLT_Anlage
• 12 Raum
• 13 Ventilator
• 14 Vorerhitzer
• 15 Wärmerückgewinnung
• 16 Zuluft
• 17 Zuluftuftfilter
• 18 Zuluftventilator

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	0.99	36	1.7080	0.4733
No log	1.99	72	0.3576	0.9228
No log	2.99	108	0.1345	0.9689
No log	3.99	144	0.0519	0.9883
No log	4.99	180	0.0410	0.9917
No log	5.99	216	0.0633	0.9881
No log	6.99	252	0.0398	0.9932
No log	7.99	288	0.0415	0.9897
No log	8.99	324	0.0424	0.9901
No log	9.99	360	0.0381	0.9901

Framework versions

• Transformers 4.22.2
• Pytorch 1.12.1+cu113
• Datasets 2.5.1
• Tokenizers 0.12.1