bert_labse-finetuning-unhealthyConv-dropout005-epochs-10

This model is a fine-tuned version of old_models/LaBSE/0_Transformer on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7820
• Mse: 0.7820
• Rmse: 0.8843
• Mae: 0.4988
• R2: 0.8587

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: 16
• eval_batch_size: 16
• seed: 42
• 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	Mse	Rmse	Mae	R2
1.5159	1.0	3389	1.2618	1.2618	1.1233	0.7882	0.7720
1.0247	2.0	6778	1.1135	1.1135	1.0552	0.7067	0.7988
0.7849	3.0	10167	1.1353	1.1353	1.0655	0.7289	0.7949
0.6271	4.0	13556	0.9255	0.9255	0.9620	0.6331	0.8328
0.5029	5.0	16945	0.9135	0.9135	0.9558	0.6148	0.8349
0.3947	6.0	20334	0.8166	0.8166	0.9036	0.5446	0.8525
0.3264	7.0	23723	0.8280	0.8280	0.9099	0.5552	0.8504
0.2774	8.0	27112	0.8125	0.8125	0.9014	0.5408	0.8532
0.2245	9.0	30501	0.7870	0.7870	0.8871	0.5034	0.8578
0.2028	10.0	33890	0.7820	0.7820	0.8843	0.4988	0.8587

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3