BACnet-Klassifizierung-Kaeltettechnik-bert-base-german-cased

This model is a fine-tuned version of bert-base-german-cased on the gart-labor "klassifizierung_kaelte_v2" dataset. It achieves the following results on the evaluation set:

• Loss: 0.0466
• F1: [0.85714286 0.98507463 1. 1. ]

Model description

This model makes it possible to classify the refrigeration components described with the BACnet standard into different categories. The model is based on a German-language data set.

Intended uses & limitations

The model divides descriptive texts into the following refrigeration categories: Free_Cooling, Refrigeration_General, Chiller, Cold Storage and Recooling Plant

Training and evaluation data

The model is based on a German-language data set.

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 8.0

Training results

Training Loss	Epoch	Step	Validation Loss	F1
0.0426	0.85	5	0.0439	[0.85714286 0.98507463 1. 1. ]
0.0175	1.85	10	0.0466	[0.85714286 0.98507463 1. 1. ]

Framework versions

• Transformers 4.21.1
• Pytorch 1.12.0+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1