distilbert-base-multilingual-cased_regression_finetuned_news_all

This model is a fine-tuned version of distilbert/distilbert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8565
• Mse: 0.8565
• Mae: 0.5566
• Rmse: 0.9255
• Mape: inf
• R Squared: 0.5011

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 2175
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Mse	Mae	Rmse	Mape	R Squared
0.9354	1.0	21755	0.9298	0.9298	0.5985	0.9643	inf	0.4584
0.8432	2.0	43510	0.8988	0.8988	0.5756	0.9481	inf	0.4764
0.8033	3.0	65265	0.8810	0.8810	0.5685	0.9386	inf	0.4868
0.8119	4.0	87020	0.8778	0.8778	0.5623	0.9369	inf	0.4887
0.7401	5.0	108775	0.8565	0.8565	0.5566	0.9255	inf	0.5011
0.6964	6.0	130530	0.8877	0.8877	0.5587	0.9422	inf	0.4829
0.6213	7.0	152285	0.8918	0.8918	0.5607	0.9444	inf	0.4805

Framework versions

• Transformers 4.39.3
• Pytorch 2.2.1
• Datasets 2.18.0
• Tokenizers 0.15.2